American Gastroenterological Association Institute Technical Review on the Role of Therapeutic Drug Monitoring in the Management of Inflammatory Bowel Diseases

Gastroenterology 2017;-:1–23

Q1

Q12

American Gastroenterological Association Technical Review on the Role of Therapeutic Drug Monitoring in the Management of Inflammatory Bowel Diseases Niels Vande Casteele,1 Hans Herfarth,2 Jeffry Katz,3 Yngve Falck-Ytter,3,4 and Siddharth Singh1 1

Division of Gastroenterology, University of California, San Diego, La Jolla, California; 2Division of Gastroenterology and Hepatology, University of North Carolina, Chapel Hill, North Carolina; 3Division of Gastroenterology and Liver Disease, Case Western Reserve University, Cleveland, Ohio; and 4VA Medical Center, Cleveland, Ohio Therapeutic drug monitoring (TDM), which involves measurement of drug or active metabolite levels and antidrug antibodies, is a promising strategy that can be used to optimize inflammatory bowel disease therapeutics. It is based on the premise that there is a relationship between drug exposure and outcomes, and that considerable interindividual variability exists in how patients metabolize the drug (pharmacokinetics) and the magnitude and duration of response to therapy (pharmacodynamics). Therefore, the American Gastroenterological Association has prioritized clinical guidelines on the role of TDM in the management of inflammatory bowel disease. To inform these clinical guidelines, this technical review was developed in accordance with the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) framework for interventional and prognostic studies, and focused on the application of TDM for biologic therapy, specifically anti-tumor necrosis factorLa agents, and for thiopurines. Focused questions address the benefits and risks of a strategy of reactive TDM (in patients with active inflammatory bowel disease) to guide treatment changes compared with empiric treatment changes, and the benefits and risks of a strategy of routine proactive TDM (during routine clinical care in patients with quiescent disease) compared with no routine TDM. Additionally, the review addresses the benefits and risks of routine measurement of thiopurine methyltransferase enzyme activity or genotype before starting thiopurine therapy compared with empiric weight-based dosing and explores the performance of different trough drug concentrations for antiLtumor necrosis factor agents and thiopurines to inform clinical decision making when applying TDM in a reactive setting. Due to a paucity of data, this review does not address the role of TDM for more recently approved biologic agents, such as vedolizumab or ustekinumab.

D

uring the last decade, the approach to treating inflammatory bowel diseases (IBD) has evolved from controlling symptoms and achieving clinical remission to decreasing progressive bowel damage and disability through the timely and optimal use of biologic therapies and/or immunomodulator agents. One emerging strategy in optimizing the use of biologics is therapeutic drug monitoring (TDM), which involves measuring serum drug

concentration (typically at trough) and anti-drug antibodies (ADAbs).1 The proposed rationale for TDM is that a systematic and algorithmic assessment of drug concentration (and ADAb) can help objectively evaluate potential reasons for failure of therapy and define next steps in management, and proactively provide opportunities for optimizing therapy to maximize chances of treatment success. This is based on clinical observations, including the presence of an exposure
response relationship in which serum drug concentration determines the magnitude of the clinical response; inter-individual variability in drug clearance through both immune-mediated (formation of neutralizing ADAb) and non
immune-mediated mechanisms (associated with high inflammatory burden), which contribute to differences in drug concentration; and concept of mechanistic failure, in which, despite adequate drug exposure at site of receptor, some patients may not respond to a particular class of biologics due to differences in underlying disease pathophysiology.2–4 A similar concept of TDM also applies to thiopurines, wherein thiopurine methyltransferase (TPMT) enzyme activity (or genotype) influences drug metabolism and concentration of drug metabolites, 6-thioguanine (6-TGN) and 6-methylmercaptopurine (6-MMP), which have been variably associated with drug efficacy and safety.5,6 However, despite increasing adoption of TDM in clinical practice, there is limited synthesis of evidence and a lack of guidance on the benefits, risks, and overall approach to TDM in the management of IBD. Therefore, the American Gastroenterological Association (AGA) has prioritized this topic for the generation of clinical guidelines.

Abbreviations used in this paper: ADAb, anti-drug antibody; AGA, American Gastroenterological Association; AZA, azathioprine; CBC, complete blood count; CD, Crohn’s disease; CI, confidence interval; CRP, C-reactive protein; ECLIA, electrochemiluminescence immunoassay; ELISA, enzyme-linked immunosorbent assay; GRADE, Grading of Recommendations Assessment, Development, and Evaluation; HMSA, homogenous mobility shift assay; IBD, inflammatory bowel disease; 6-MMP, 6-methylmercaptopurine; 6-MP, 6-mercaptopurine; RCT, randomized controlled trial; RIA, radioimmunoassay; RR, relative risk; TDM, therapeutic drug monitoring; 6-TGN, 6-thioguanine; TNF, tumor necrosis factor; TPMT, thiopurine methyltransferase; UC, ulcerative colitis. © 2017 by the AGA Institute 0016-5085/$36.00 http://dx.doi.org/10.1053/j.gastro.2017.07.031

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

AGA SECTION

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120

2

Gastroenterology Vol.

Objectives of the Review This technical review addresses the following focused clinical questions on different strategies of TDM with biologics and thiopurines to improve patient outcomes: 1. In biologic-treated patients with active IBD, what are the benefits and risks of reactive TDM (in response to active disease) to guide treatment decisions over a strategy of empiric treatment changes? If TDM is adopted, what is the association between different target drug concentrations and clinical outcomes? 2. In biologic-treated patients with quiescent IBD, what are the benefits and risks of routine proactive TDMguided dose adaptation? 3. What are the benefits and risks of routine measurement of TPMT enzyme activity or genotype, before starting thiopurines, over empiric weight-based dosing? 4. In thiopurine-treated patients with active IBD or suspected to have thiopurine-related toxicity, what are the benefits and risks of reactive TDM with measurement of 6-TGN and 6-MMP levels, to guide treatment decisions over a strategy of empiric treatment changes? If TDM is adopted, what target 6-TGN cutoff is optimal for improving clinical outcomes? 5. In thiopurine-treated patients with IBD on standard weight-based therapy, what are the benefits and risks of routine proactive TDM-guided dose adaptation? The results of this technical review were used to inform the development of the accompanying clinical guidelines on TDM in IBD. Of note, we focused on anti-tumor necrosis factor (TNF)-a agents only when reviewing TDM for biologics, and no distinction was made between monotherapy and combination therapy in this setting. While the same concepts may apply to other biologics (eg, vedolizumab and ustekinumab), their mechanism of action is distinct from anti-TNF agents and, at this point, there are very limited published data on the role of TDM for these agents to inform guidelines. Therefore, non
anti-TNF biologics are not discussed in these guidelines. Similarly, due to limited use, the technical review team and the guideline panel, with the approval of the AGA Governing Board, opted not to synthesize evidence on the role of TDM for methotrexate, cyclosporine, and tacrolimus. AGA SECTION

121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180

Casteele et al

Methods Overview This technical review and the accompanying guideline were developed using the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) framework.7 The members of the technical review panel were selected by the AGA Clinical Guidelines Committee based on their clinical content and guidelines methodological expertise and went through a thorough vetting process for potential conflicts of interest. Through an iterative process, and in conjunction with

-,

No.

-

the guideline panel, the participants developed focused clinical questions on the role of TDM for anti-TNF agents and thiopurines deemed relevant for clinical practice that the guideline would address. After the focused questions were approved by the AGA Governing Board (in December 2015), the technical review team formulated the clinical questions, identified relevant patient-important outcomes, systematically reviewed and summarized the evidence for each outcome across studies, and then rated the quality of the evidence across all outcomes for each clinical question.

Formulation of Clinical Questions Using the PICO format, which frames a clinical question by defining a specific population (P), intervention (I), comparator (C), and outcomes (O), the team finalized 5 questions (Table 1). The first set of PICOs focused on TDM for anti-TNF agents, and the second set of PICOs on TDM for thiopurines. Questions focused on comparing different strategies of TDM classified as reactive TDM or routine proactive TDM. Reactive TDM is defined as TDM performed in response to active IBD (ongoing active inflammation based on biochemical, endoscopic, or radiologic assessment, usually with symptoms) after a period of quiescent disease, or continued inflammation without achieving remission with index therapy; of note, a small fraction of patients, especially those with active Crohn’s disease (CD) (active inflammation) may be asymptomatic, and the concept of reactive TDM also applies to those patients. Routine proactive TDM was defined as TDM performed in patients regardless of clinical status (generally in quiescent disease) periodically as part of routine clinical care. The comparator strategy relied on empiric treatment changes—for anti-TNF agents, this focused on a stepwise approach of empiric escalation of therapy or switching to different treatment agents within or outside the index class (ie, with same putative mechanism of action or with a different mechanism of action). Potentially relevant patient-important outcomes were considered and rated in terms of importance through consensus; clinical remission was considered critical for decision making, whereas mucosal healing (endoscopic remission), serious adverse events, cost, drug or metabolite concentration, and patient convenience were considered important outcomes. The panel recognized limitations of using a clinical disease activity as an outcome measure, especially for CD, but still believed that in the current context, it is the most consistently reported outcome in clinical practice and is important for patients.

Search Strategy and Study Selection Criteria The literature search was performed on March 6, 2016, and details of the search strategy are reported in the Supplementary Material. Studies were selected for inclusion based on PICO theme. Due to lack of high-quality randomized controlled trials (RCTs) informing each question, the study selection and data synthesis approach were customized for each question. For PICO #1 (reactive TDM) and PICO #2 (routine proactive TDM), for anti-TNF agents, we included RCTs, comparative observational studies, or cohort studies in adults with IBD, with either active IBD or quiescent disease, treated with anti-TNF agents, who underwent TDM (ie, measurement of drug levels and/or ADAbs). Due to the paucity of high-quality RCTs and observational comparative studies for

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240

241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 -

2017

Table 1.Focused Clinical Questions and Corresponding Questions in PICO (Population, Intervention, Comparator, and Outcomes) Format Addressed in This Technical Reviewa PICO question Focused question

Thiopurine therapy in IBD 3. In patients with IBD being started on thiopurines, is routine TPMT enzyme activity or genotype measurement (to guide dosing) superior to no TPMT measurement (with empiric weight-based dosing of thiopurines)? 4. In patients with active IBD treated with thiopurines or with side effects thought to be due to thiopurine toxicity, is reactive TDM to guide treatment changes superior to no TDM with empiric treatment changes? 5. In patients with IBD treated with thiopurines, is routine TDM to guide thiopurine dosing superior to empiric weight-based dosing?

Intervention

Comparator

Outcomes

Adults with active IBD, being treated with anti-TNF agents

Reactive TDM (measurement of trough and anti-drug antibodies), in response to new or continuing active IBD), to guide treatment changes Routine proactive TDM (measurement of trough drug level and anti-drug antibodies), to guide ongoing treatment, regardless of clinical status

No reactive TDM, with either empiric dose escalation, or switching therapy

Clinical remission or response

No routine TDM, in absence of change in clinical status

Clinical remission or response

Adults with quiescent IBD, being treated with anti-TNF agents

Adults with IBD, starting thiopurine therapy

Routine TPMT measurement (enzyme activity or genotype) to guide thiopurine dosing

No TPMT measurement, with empiric weight-based thiopurine dosing

Serious adverse events

Adults with active IBD, being treated with thiopurines, or with side effects concerning for thiopurine toxicity

Reactive TDM (measurement of 6TGN and 6-MMP), in response to active IBD or side effects concerning for thiopurine toxicity, to guide treatment changes Routine proactive TDM (measurement of 6-TGN and 6-MMP), to guide ongoing treatment, regardless of clinical status

No reactive TDM, with empiric treatment changes

Clinical remission or response Serious adverse events

No routine TDM, with empiric weight-based thiopurine dosing

Clinical remission or response Serious adverse events

Adults with IBD, being treated with thiopurines

a

Besides these questions, the technical review also addresses the performance of different trough levels for infliximab, adalimumab, certolizumab pegol, and golimumab in clinical practice to inform PICOs #1 and #2.

AGA Review of Therapeutic Drug Monitoring in IBD

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

Biologic therapy in IBD 1. In adults with active IBD treated with anti-TNF agents, is reactive TDM to guide treatment changes superior to no TDM and empirically escalating dose or switching therapies? 2. In adults with quiescent IBD treated with anti-TNF agents, is routine proactive TDM to guide prospective treatment changes superior to no TDM?

Population

3

AGA SECTION

301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360

4

Gastroenterology Vol.

PICOs #1 and #2, we relied on cohort studies that reported differences in outcomes of patients depending on trough level and/or presence of ADAb, in response to empiric dose escalation and/or switching therapies. This provided indirect evidence on potential risks and benefits of TDM-guided treatment decisions compared with empiric treatment changes. For PICO #3 on the role of TPMT enzyme activity or genotype before starting thiopurines, we included RCTs in patients with IBD who were started on thiopurines, based on either TPMT guidance or empirically, and evaluated safety and efficacy of therapy. For PICOs #4 and #5 on application of TDM strategies for thiopurines, a similar approach was adopted, wherein, if highquality RCTs or observational comparative studies were lacking, cohort studies were used to inform evidence indirectly. To inform optimal target trough concentrations and their performance, we initially searched for RCTs or observational comparative studies that report differences in patient outcomes based on different target trough concentration thresholds. In the absence of comparative studies, we chose cohort and crosssectional studies that reported correlation between different thresholds and presence or absence of clinical remission (or response) and assessed the pooled proportion of patients “not in remission” above certain predefined thresholds. Of note, these were not framed as PICOs, but are rather presented semiquantitatively to inform clinical guidelines.

Statistical Analysis When comparative studies were available, pooled relative risk (RR) and 95% confidence intervals (CI) were calculated using DerSimonian-Liard random-effects model.8 For PICOs #1 and #4, as well as in assessing performance of different trough concentration thresholds, we estimated weighted pooled proportion of patients in different relevant categories, using random-effects meta-analysis. Statistical heterogeneity was assessed using the I2 statistic, and values >50% were considered suggestive of significant heterogeneity.9 Small study effects were examined using funnel plot symmetry and Egger’s regression test, though it is important to recognize that these tests are unreliable when the number of studies is <10.10 Statistical analyses were performed using RevMan, version 5.3 (Cochrane Collaboration, Copenhagen, Denmark) or Comprehensive Meta-Analysis software, version 2 (Biostat, Englewood, NJ).

Quality of Evidence

AGA SECTION

361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420

Casteele et al

The GRADE approach was used to rate the quality of evidence (or confidence in summary effect estimates).7 In this approach, direct evidence from RCTs starts at high quality and can be rated down based on risk of bias in the body of evidence (or study quality), indirectness (addressing a different but related population, intervention, or outcome, from the one of interest), imprecision (of summary estimate and boundaries of 95% CI), inconsistency (or heterogeneity), and/or publication bias to levels of moderate, low, and very low quality. Due to inherent limitations in observational studies (ie, selection bias, unmeasured confounding), evidence derived from observational studies starts at low quality, and is then potentially downgraded based on the factors mentioned, or can be upgraded in case of dose
response relationship and large magnitude of effect.

-,

No.

-

Evidence-to-Decision Framework Because this technical review was used to inform the development of clinical guidelines, besides a comprehensive risk
benefit analysis and the accompanying quality of evidence, information about additional factors, such as patients’ values and preferences, cost-effectiveness, and resource utilization were also considered.

Primer on Therapeutic Drug Monitoring With AntiLTumor Necrosis Factor Agents Biologics, more specifically monoclonal antibodies, are protein-based drugs that can bind to an antigen with great specificity and possess powerful effector functions. After parenteral administration (subcutaneous or intravenous), monoclonal antibodies are distributed mainly in the central compartment and are eliminated through proteolytic catabolism after receptor-mediated endocytosis in the cells of the reticuloendothelial system.11 Unique for biologics is that the immune system can recognize these molecules as non-self, leading to a humoral or cell-mediated immune response with the formation of ADAb.12 ADAb impair the efficacy of the drug by either blocking the antigen-binding site or by forming complexes with the molecule, leading to up-regulated clearance. Because of the formation of ADAb and changes in other covariates, such as body mass, albumin, C-reactive protein (CRP), and others, considerable inter-individual variability in drug exposure is observed.13 The drug concentration at the site of the receptor determines the magnitude of the pharmacologic response and monitoring drug exposure in individual patients might be helpful to guide therapeutic decisions in clinical practice. Various measures can be used for drug exposure, but serum trough concentrations, or the concentration of drug just before the next administration, is used most commonly. Assessing the exposure
response relationship in individual patients can be done by TDM, with measurement of trough concentrations at various time points. Because of different dosing regimens during induction vs maintenance therapy, ideal thresholds where the exposure
response relationship is optimal are different in different phases of therapy. For example, patients may show impaired response to therapy due to pharmacokinetic failure because of inadequate drug exposure as a result of increased drug clearance. This is distinct from the scenario where patients have adequate drug exposure but show no response to therapy due to mechanistic failure because the disease is not mediated by the antigen that is targeted by the molecule.14 In the former scenario, dose escalation may result in recaptured response, and in the latter scenario, switching to a drug out of class with another mechanism of action might be the preferred pharmacologic strategy.

Assays for Therapeutic Drug Monitoring of Anti
Tumor Necrosis Factor Agents Different assay formats exist to measure drug and ADAb, of which the most commonly used are the enzyme-linked

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480

481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540

Q6

2017

AGA Review of Therapeutic Drug Monitoring in IBD

immunosorbent assay (ELISA), homogenous mobility shift assay (HMSA), and electrochemiluminescence immunoassay (ECLIA).15 The main characteristics of the assays used for anti-TNF currently available in the United States are outlined in Supplementary Table 1. ADAb assays that are carried out in a fluid phase environment (HMSA, ECLIA, and radioimmunoassay [RIA]) are more sensitive to detect lowaffinity antibodies than solid-phase ADAb assays (ELISA). For measuring ADAbs, no international analytical standard is currently available and different assays report different ADAb titers (eg, mg/m, mg/mL equivalents, U/mL, and AU [arbitrary units]/mL). In general, measurement of infliximab concentrations with various assays was found to be relatively comparable with acceptable specificity, accuracy, and reproducibility between assays. This was shown in a recent large comparative study evaluating various assays for the measurement of infliximab.15 Agreement between assay methods was quantified by the intra-class correlation coefficient of the Janssen infliximab assay (ELISA) with the assay from KU Leuven (ELISA), Sanquin (ELISA), Dynacare (ELISA), and Esoterix/LabCorp (ECLIA) and was 0.960, 0.895, 0.931, and 0.971, respectively. Infliximab recovery of a sample of pooled healthy control serum spiked with 5 mg/mL infliximab ranged between
7% and þ20% of the nominal concentration, which was determined by the authors to be within the acceptable range.16,17 In the same comparative study, ADAbs to infliximab were detected specifically and reproducibly.15 ADAbs to infliximab methods of Janssen (ECLIA), Sanquin (RIA), and Esoterix/LabCorp (ECLIA) were more resistant to infliximab interference than Dynacare (ELISA) and KU Leuven (ELISA), which were affected by infliximab concentrations of 2 mg/mL. Although the correlation between ADAb assays might be reasonable, comparing the quantitative result between assays is less ideal, that is, although quantitative ADAbs detected using HMSA, ELISA, or ECLIA will be very different, their results should be interpreted in the appropriate clinical context and drug concentration levels (the latter would be quantitatively comparable across different assays). A comparison of 3 assays for the detection of ADAb to infliximab (bridging ELISA, RIA, and HMSA) showed that overall, assays provided similar guidance for clinical practice in most patients with loss of response.18 For golimumab, 2 comparative studies evaluated 2 commercially available tests (Promonitor [ELISA] and Sanquin [ELISA]) for the detection of drug in serum, which showed excellent correlation and agreement.19 For adalimumab, 1 comparative study evaluated a commercially available ELISA (Matriks Biotek, Ankara, Turkey) and HMSA (Prometheus Laboratories, Inc, San Diego, CA) for the detection of drug in serum in 23 patients treated with maintenance adalimumab during a 96-week follow-up period.20 HMSA measured consistently lower concentrations than ELISA, resulting in a 2-fold difference in adalimumab trough cutoffs that correlated with clinical remission. These data warrant caution when extrapolating cutoffs across assays, especially for adalimumab. Currently, data on the comparison of assays for certolizumab pegol are lacking.

5

Interpreting Results of Therapeutic Drug Monitoring All available drug assays report both drug concentration and ADAbs. In interpreting these, we suggest evaluating drug concentration first. In the presence of therapeutic drug concentration, ADAbs are generally undetectable. If detected, ADAbs are generally low concentration and, consequently, can have a limited effect on drug pharmacokinetics. In a combined retrospective analysis of 4 studies, it was observed that both infliximab concentration (odds ratio, 1.8; 95% CI, 1.3
2.5) and ADAbs to infliximab (odds ratio, 0.57; 95% CI, 0.39
0.81) were independently associated with biochemical remission (CRP 5 mg/L) in patients with CD on maintenance infliximab therapy.21 Detecting both infliximab and ADAbs to infliximab in the same sample was observed in a minority of samples (7.3%), suggesting that impaired efficacy is likely a result of both a mechanism of accelerated drug pharmacokinetics and direct blocking of drug activity by ADAbs. In patients who are failing despite a therapeutic drug concentration, regardless of ADAb, this can be caused by mechanistic failure, and patients might benefit from switching to a drug out of class. On the other hand, if drug concentration is subtherapeutic or undetectable, then we suggest evaluating ADAbs. If ADAbs are detectable, particularly at high titers, then it suggests that drug clearance is increased due to immune-mediated mechanisms and switching to a drug in class with the same mechanism of action but other molecular structure may be the preferred pharmacologic option. If ADAb are not detectable, then clearance is up-regulated due to non
immune-mediated mechanisms, such as high inflammatory burden resulting in rapid drug utilization and/or excessive drug wasting due to fecal loss (indicated by elevated CRP and/or fecal calprotectin and/or low albumin).13,22 Table 2 summarizes the conventionally proposed interventions based on TDM in the reactive setting in patients with secondary loss of response, and the estimated distribution of patients in each category, based on 3 observational studies.23–25 Most importantly, these possible interventions at time of secondary loss of response remain to be confirmed in a well-designed prospective study.

Results Question 1: In Adults With Active Inflammatory Bowel Disease Treated With Anti
Tumor Necrosis Factor Agents, Is Reactive Therapeutic Drug Monitoring to Guide Treatment Changes Superior to No Therapeutic Drug Monitoring and Empirically Escalating Dose or Switching Therapies? Key message. In patients with active IBD treated with anti-TNF agents, there may be a benefit of reactive TDM to guide treatment changes over empirically escalating dose or switching therapies (very-low-quality evidence). Effect estimates. One small RCT26 and 3 observational studies23–25 in patients with IBD on maintenance therapy

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

AGA SECTION

-

541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600

6

Gastroenterology Vol.

-,

No.

-

Table 2.Proposed But Not Yet Prospectively Validated Interventions Based on Therapeutic Drug Monitoring in the Reactive Setting in Patients With Secondary Loss of Response, and Observed Distribution of Patients in Different Categories Based on Observational Studies23–25 Drug concentration ADAb

Subtherapeutic drug trough concentration

Undetectable ADAbs

Nonimmune-mediated pharmacokinetic failure, 51% Y Dose escalate by either increasing the dose or decreasing the interval between drug administrations Immune-mediated pharmacokinetic failure, 19% Y Consider optimizing index therapy in case of low-level ADAbs, or switch to drug in class and consider adding an immunomodulator in patients with high level ADAbs

Detectable ADAbs

AGA SECTION

601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660

Casteele et al

with anti-TNF agents were used to inform this question. An additional RCT designed to address this question was also identified on ClinicalTrials.gov (Number: NCT01960426). However, that trial was terminated very shortly after start due to difficulties with recruitment. In a Danish RCT, 69 patients with CD on maintenance therapy with infliximab who developed CD-related symptoms were randomized to either TDM-guided treatment changes (based on trough concentration and presence of antibodies to infliximab; n ¼ 33) vs empiric dose escalation (5 mg/kg every 4 weeks; n ¼ 36).26 In this trial, infliximab and ADAbs to infliximab were measured by fluid-phase RIA.27 Infliximab trough concentration was classified as therapeutic (0.5 mg/mL) or subtherapeutic (<0.5 mg/mL), and ADAb to infliximab as detectable or undetectable (limit of quantification, 10 U/mL). On intention-to-treat analysis 12 weeks after intervention, 10 of 33 patients (30.3%) in the TDM-guided treatment group vs 14 of 36 (38.9%) in the empiric dose-escalation arm achieved clinical remission (RR of achieving clinical remission, 0.78; 95% CI, 0.40
1.51). Based on the investigator’s criteria, the majority of patients in the TDM-guided dosing arm (76%) had a therapeutic infliximab trough (0.5 mg/mL) at the time of secondary loss of response, suggesting mechanistic failure, and these patients were treated with discontinuation of infliximab and switching out of biologic class (non
antiTNF-mediated therapy), or to corticosteroids and/or immunomodulators, or by targeting potential noninflammatory causes of symptoms. Treatment failure was presumably due to immunogenicity of infliximab in 20% of patients (infliximab trough concentrations <0.5 mg/mL with detectable ADAbs to infliximab), who were treated with switching within class (to an alternate anti-TNF agent), and only 4% of patients had non
immune-mediated pharmacokinetic failure (with infliximab trough concentrations <0.5 mg/mL and undetectable ADAbs to infliximab), who were managed with infliximab dose escalation to 5 mg/kg every 4 weeks. In contrast, all patients randomized to

Therapeutic drug trough concentration Mechanistic failure, 25% Y Switch to drug out of class

Mechanistic failure, 5% Y Switch to drug out of class and consider adding an immunomodulator in case of switching to a biologic

non
TDM-guided dosing underwent empiric dose escalation at time of secondary loss of response. Three observational studies also contributed to the evidence on the role of reactive TDM after secondary loss of response in patients on maintenance anti-TNF therapy.23–25 These studies are summarized in Supplementary Table 2. In these studies, all patients with secondary loss of response underwent empiric dose escalation as standard of care. Subsequently, difference in response to dose escalation was retrospectively assessed in patients in different strata based on anti-TNF trough concentration and presence or absence of ADAbs, that is, what proportion of patients with therapeutic trough without ADAb (mechanistic failure), subtherapeutic trough without ADAbs (non
immune-mediated pharmacokinetic failure), and subtherapeutic trough with ADAbs (immune-mediated pharmacokinetic failure), responded to empiric escalation of therapy. On pooled analysis of 2 studies, overall 60 of 134 (45%) patients responded to dose escalation.23,25 On evaluating by different strata, 41 of 50 (82%) patients with non
immune-mediated pharmacokinetic failure responded to dose escalation (RR of achieving response with dose escalation vs empiric dose escalation, 1.71; 95% CI, 1.39
2.11). In contrast, only 2 of 24 (8%) patients with immune-mediated pharmacokinetic failure (with high ADAbs) responded to dose escalation (RR vs empiric dose escalation, 0.26; 95% CI, 0.08
0.86). Yanai and colleagues24 observed similar results—55% of loss of response events due to non
immune-mediated pharmacokinetic failure (no/low ADAbs) vs 15% events due to immune-mediated pharmacokinetic failure (high ADAbs) responded to dose escalation with index agent. In contrast to the RCT by Steenholdt et al,26 only 139 of 464 (30%) patients/loss of response events in observational studies were classified as being due to mechanistic failure (vs 76% in the RCT), partly due to higher trough concentration thresholds (infliximab trough, 2.0
3.8 mg/ mL; adalimumab trough, 4.5
4.9 mg/mL) (Table 2).26 The majority (235 of 464 [51%]) of patients in these studies

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720

721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780

2017

AGA Review of Therapeutic Drug Monitoring in IBD

were deemed to have secondary loss of response due to non
immune-mediated pharmacokinetic failure (subtherapeutic trough without ADAbs) (vs 4% in RCT), and 19% (90 of 464) of patients were classified as having immunemediated pharmacokinetic failure (subtherapeutic trough with high ADAbs) (vs 20% in RCT). Besides dose escalation, an alternative empiric response to secondary loss of response may be switching to an alternative agent (either within class with same mechanism of action, or outside of class with a different mechanism of action). Roblin et al25 observed that 16 of 52 (31%) patients responded to empiric switching to a different anti-TNF after failure of empiric dose escalation. On retrospectively applying TDM, only 2 of 29 (7%) patients with therapeutic trough with index anti-TNF responded to switching to a different anti-TNF agent. Among patients with low trough, 12 of 15 (80%) patients with immune-mediated pharmacokinetic failure (high ADAbs) responded to switching to another anti-TNF agent, whereas only 2 of 8 (25%) patients with non
immune-mediated pharmacokinetic failure responded to switching to a different anti-TNF agent after failure to respond to empiric dose escalation. Yanai et al24 made similar observations—among patients with high ADAbs at time of loss of response, the likelihood of response to switching to different anti-TNF was much higher compared with dose escalation of index agent (6 months after intervention, 50% vs 15%; P ¼ .03). Among patients with low ADAbs at time of loss of response, likelihood of response to dose escalation was higher compared with switching to different anti-TNF (6 months after intervention, 50% vs 40%; P ¼ .02).24 Following the TDM framework, they observed that among patients with therapeutic trough at time of secondary loss of response (mechanistic failure), likelihood of response to switching out of biologic class (to a non
anti-TNF-mediated mechanism) was numerically higher compared with attempts at anti-TNF optimization (based on ADAb level, either dose escalation or switching within class) (6 months after intervention, 78% vs 44%; P ¼ .09). Quality of evidence. The single RCT that informed this question was at high risk of bias, with a high degree of nonadherence to protocol (only 19 of 33 patients randomized to TDM-guided intervention were handled accordingly, and 17 completed the trial per protocol). The threshold of therapeutic infliximab trough used in this trial (0.5 mg/mL) is probably too low (and a significant proportion of patients may have been incorrectly labeled as having mechanistic failure); the distribution of patients in terms of TDM-defined reasons for treatment failure (and subsequent intervention) was very different from what is observed in real-world observational studies, resulting in indirectness. Additionally, the summary estimate was very imprecise, with very wide CIs. In contrast to RCT-level evidence, observational studies suggest a significant difference in likelihood of response to empiric dose escalation (or switching to different therapy), depending on which TDM-defined stratum a patient falls into. Due to the observational nature of these studies with inherent risks of bias, and low event rates in included

7

studies (resulting in imprecision), the quality of evidence derived from observational studies was rated as very low. Therefore, the overall body of evidence supporting the use of reactive TDM to guide treatment changes over empiric dose escalation or switching therapies in patients with IBD treated with anti-TNFs with secondary loss of response is very low. Potential harms of intervention. TDM is performed through a blood test generally just before next due dose, which can cause a small inconvenience. However, there are additional potential harms of downstream interventions from TDM testing because thresholds for therapeutic trough concentration and ADAbs are not very well-defined. Therapeutic trough is dynamic, depending on phase of intervention (induction vs maintenance therapy), treatment target (clinical remission vs endoscopic remission), and phase of disease activity (severe active vs mild active disease; luminal disease vs perianal disease).28–31 Likewise, threshold ADAb levels that define immunogenicity and predict low likelihood of response to dose escalation are not well-defined and are variable across assays. Therefore, due to variable thresholds, which have suboptimal discriminatory performance, strict adherence to TDM-guided treatment changes can potentially result in inappropriate treatment changes in some patients who might have responded to empiric escalation of therapy. In contrast, empiric escalation of therapy can impose significant additional costs of escalated therapy, result in potentially futile therapy, and delay more effective therapy, particularly if repeated cycles of empiric dose escalation are attempted. In addition, in patients with immune-mediated pharmacokinetic failure (with established ADAbs), additional drug exposure is likely to result in immediate and delayed hypersensitivity reactions, which are generally mild but can be severe. Similarly, excessive drug exposure can conceivably result in higher risk of drug-related adverse events, such as serious infections, although this has not been consistently observed. Discussion. Approximately one third of IBD patients fail to respond to anti-TNF therapy and, of those who do respond initially, secondary loss of response is an important clinical problem and occurs in up to 40% of patients during the first year of therapy.32,33 Once noninflammatory complications, such as strictures and overlapping irritable bowel syndrome, are excluded, potential causes of this phenomenon include low drug exposure because of immune- or non
immune-mediated clearance mechanisms, or mechanistic failure whereby non
TNF-mediated cytokine pathways may be activated. Given the complexity of this problem, it is hypothesized that TDM may be helpful to guide appropriate treatment changes. However, based on current evidence, the magnitude of benefit of this approach over empiric dose escalation is uncertain. Several factors contribute to this uncertainty. First, the included studies were at high risk of bias, with poor adherence to protocol, and inconsistent and often retrospectively defined thresholds for optimal trough and ADAbs. Second, most studies relied on clinical disease activity with or without biochemical or endoscopic confirmation of disease activity, to initiate

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

AGA SECTION

-

781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840

8

Gastroenterology Vol.

intervention, and to measure success of intervention. Clinical disease activity has modest correlation with endoscopic disease activity, risk of surgery, and disease-related complications, at least in patients with CD.34 Third, besides the mild inconvenience of blood test for TDM, there is potential misclassification and missed treatment optimization opportunities due to suboptimally defined thresholds for both drug concentration and ADAb. In contrast, empiric dose escalation may result in additional costs, higher likelihood of futile treatment and disease-related complications, and potentially delay more effective therapy. Overall, the risk
benefit profile may favor reactive TDM-guided treatment changes over empiric dose escalation. In the Danish RCT, in patients with secondary loss of response to infliximab, accumulated costs related to treatment of CD during a 12-week period were significantly lower (34%) for those who underwent TDM-guided therapy vs empiric dose escalation of infliximab (V6038 vs V9178, respectively; P < .001).26 In a decision analysis, Velayos et al35 reported TDM-guided treatment strategy dominated empiric dose escalation (numerically lower costs, and higher quality-adjusted life year), which was stable across multiple sensitivity analyses with variable model inputs. They estimated that increasing the cost of the test >25-fold their base-case scenario ($5700) resulted in the testing strategy no longer being less expensive than the empiric strategy ($37,267 vs $37,266). It is important to note that all studies used to synthesize this evidence included only those patients with secondary loss of response on maintenance infliximab or adalimumab, that is, patients who had responded initially to therapy and subsequently lost response during the maintenance phase of therapy. It is difficult to extrapolate results of a reactive TDM-based treatment strategy vs empiric treatment escalation to induction therapy. Recent studies in patients treated with infliximab and adalimumab suggest that suboptimal drug exposure can play a role during the induction phase, particularly in patients with a high inflammatory burden and/or individual characteristics that are associated with increased clearance, such as low albumin concentration.36,37 Knowing what thresholds are associated with outcomes shortly after starting therapy could be useful in guiding treatment decisions.

AGA SECTION

841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900

Casteele et al

Question 2. In Adults With Quiescent Inflammatory Bowel Disease Treated With Anti
Tumor Necrosis Factor Agents, Is Routine Proactive Therapeutic Drug Monitoring to Guide Prospective Treatment Changes Superior to No Therapeutic Drug Monitoring? Key message. In patients with quiescent IBD treated with anti-TNF agents, the benefit of routine proactive TDM over no therapeutic monitoring is uncertain (very-lowquality evidence). Effect estimates. We did not identify any RCT or comparative observational study evaluating the role of routine proactive TDM for achieving remission. Therefore, indirect evidence was derived from a single RCT—the

-,

No.

-

Trough Concentration Adapted Infliximab Treatment (TAXIT) trial. In TAXIT, 263 responder patients (178 CD and 85 ulcerative colitis [UC]) on maintenance infliximab were first dose optimized to reach an infliximab trough concentration within the 3
7-mg/mL range (optimization phase).38 Once infliximab trough was within the therapeutic target range (average of 2.1 optimizations needed), patients were randomized to either continued TDM-guided dosing (ie, continuous drug monitoring before each dose to stay within target trough range, n ¼ 128) or no TDM with empiric dosing based on symptoms and CRP (n ¼ 123) during the maintenance phase. The duration of the maintenance phase was 1 year after randomization and the primary end point was defined as the proportion of patients in each group in clinical (Harvey-Bradshaw index 4 for CD and partial Mayo score 2 with no individual subscore >1 for UC) and biological remission (CRP concentration of <5 mg/L). Before the optimization phase, only 121 of 275 (44%) patients were in the predefined target trough range (3
7 mg/ mL), whereas 82 (30%) had a subtherapeutic trough concentration (<3 mg/mL) (including 24 patients with undetectable trough, of which 18 were positive for ADAb). Infliximab trough concentration was >7 mg/mL in 72 patients (26.2%). In this trial, with initial optimization, in the subset of patients with initial low trough concentration, the proportion of patients with CD in clinical remission increased (pre- vs post-optimization, 28 of 43 [65%] vs 38 of 43 [88%]). However, after initial optimization, the proportion of patients achieving remission at 1 year with continued proactive routine TDM (88 of 128 [69%]) vs no TDM (81 of 123 [66%]) (and empiric escalation in case of clinical or biochemical relapse) was not different (RR, 1.04; 95% CI, 0.88
1.24) (Table 3). Quality of evidence. The single RCT used to inform this question provides only indirect evidence on the potential role of routine proactive TDM because all patients were in clinical response on maintenance infliximab therapy and before randomization were “trough-optimized,” that is, underwent proactive monitoring to achieve a therapeutic infliximab trough between 3 and 7 mg/mL. Therefore, this trial does not inform on the role of one-time dose optimization, infrequent optimization, or on early optimization in patients at high risk of increased drug clearance due to immune-mediated or non
immune-mediated pathway. Due to summary estimate near unity, and with very wide CIs, very serious imprecision was noted. There was no serious risk of bias. Therefore, the overall quality of evidence was rated as very low. Potential harms of intervention. Routine proactive TDM can impose modest inconvenience and be expensive due to repeated testing. The feasibility of such an approach for drugs administered subcutaneously every 2
4 weeks is unclear. In addition, because target thresholds are poorly defined, inconsistent, and can vary depending on individual patient or targeted outcome, proactive TDM may lead to inappropriate treatment changes and therapeutic dilemmas in patients otherwise in remission. In contrast, there are very few direct harms to patients of not performing proactive TDM. In the TAXIT trial, at the end of 1 year, a higher

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960

9

proportion of patients randomized to no TDM developed ADAbs and undetectable infliximab trough, which can conceivably put patients at risk of disease flare in the long-term. Discussion. Unfortunately, there was no direct evidence to compare the benefit and harms of routine proactive TDM vs no TDM and, therefore, we relied on indirect evidence to inform the evidence. During the optimization phase of TAXIT, all patients underwent proactive monitoring to achieve investigator-defined target infliximab concentrations, and this led to an increase in the proportion of patients in remission (with dose escalation in the subset of patients with infliximab trough concentrations <3 mg/mL at screening), and allowed a 28% reduction in drug costs in those patients with supratherapeutic drug concentrations (>7 mg/mL) and in whom the dose was reduced safely.38 While these results suggest there might be a benefit to a subset of patients for conducting one-time proactive TDM to optimize infliximab therapy during maintenance phase, there was no comparator group for this part of the study. The randomization phase of TAXIT does inform us that, after initial optimization, continued proactive TDM and concentration-based dosing at every infusion, is not superior to clinically guided dosing, for achieving clinical and biologic remission at least up to 1 year of therapy. However, current evidence does not inform on role of one-time dose optimization, infrequent optimization (eg, once or twice a year) or on early optimization in patients at high risk of increased drug clearance due to immune- or non
immunemediated pathway, comparison of proactive TDM vs reactive TDM, or the potential benefits of routine and continuous proactive TDM over no TDM on long-term outcomes (beyond 1 year). In a single-center retrospective cohort study of patients on maintenance infliximab in clinical remission, Vaughn et al39 observed that patients who undergo proactive TDM (n ¼ 48) are significantly more likely to remain on infliximab compared with patients who do not undergo proactive TDM (n ¼ 78) (at 5 years, 86% vs 52%; hazard ratio, 0.3; 95% CI, 0.1
0.6). Rates of discontinuation of infliximab therapy due to clinical flare (0% vs 19%) or infusion reaction (2% vs 7.7%) was also lower in patients undergoing proactive TDM vs no TDM. However, there were limited data on direct patient-relevant clinical outcomes, and the Q7 study included a highly selected group of patients. Post-hoc analyses of clinical trials of induction therapy and observational studies have suggested an exposure
response relationship, with patients with higher trough concentrations between weeks 4 and 14 more likely to achieve remission on follow-up compared with patients with lower trough concentrations.40–44 In a post-hoc analysis of the ACT 1
2 trials, including 242 patients with UC receiving standard infliximab induction and maintenance therapy, Adedokun et al40 observed that, compared with patients with lowest quartile of drug concentration at week 8, patients in the 3rd and 4th quartile were significantly more likely to achieve clinical remission (Q1 vs Q3 vs Q4: 26.3% vs 43.9% vs 43.1%; P < .05) and mucosal healing (45.6% vs 71.9% vs 79.3%; P < .001). A similar exposure
response

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

AGA SECTION

RR, risk ratio. a All patients were initially “optimized” to a target trough concentration before subsequent randomization to routine proactive TDM or no TDM, and routine proactive TDM was performed at every infusion, rather than periodically. b Summary estimate near unity, and wide CIs.

26 more per 1000(79 fewer to 158 more) 659 per 1000 1.04 (0.88
1.24) 88/128 (68.8) 81/123 (65.9) 4 very low None Seriousb Very seriousa Not serious Not serious Clinical remission (critical outcome) 251 (1 RCT)

Risk difference with routine proactive TDM Risk with no TDM No. of participants (studies), follow-up

Risk of bias

With routine Overall proactive Publication quality of With no TDM TDM evidence Inconsistency Indirectness Imprecision bias

Relative effect, RR (95% CI)

Anticipated absolute effects Study event rates, n (%)

Summary of findings

AGA Review of Therapeutic Drug Monitoring in IBD

Quality assessment

961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020

2017

Table 3.In Adults with Quiescent Inflammatory Bowel Disease Treated With Anti
Tumor Necrosis Factor Agents, Is Routine Proactive Therapeutic Drug Monitoring to Guide Prospective Treatment Changes Superior to No Therapeutic Drug Monitoring?38

-

1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080

10

AGA SECTION

1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140

Casteele et al

Gastroenterology Vol.

relationship was observed at 6 weeks, on post-hoc analyses of trials of golimumab in patients with UC.45 Conceivably, there may be a benefit of early optimization of therapy during induction through proactive TDM. But there is a lack of prospective interventional data assessing the value of proactive TDM during induction to increase remission rates by targeting threshold drug concentrations.

Threshold Trough Concentrations of AntiLTumor Necrosis Factor Agents The choice of therapeutic threshold drug trough concentration is critical to understand why patients lose response to therapy and to guide treatment decisions. It is important to note that target trough concentrations are not absolute values, but rather represent a spectrum with an exposure
response relationship, and may vary by treatment target (clinical remission vs endoscopic remission), disease state (high disease activity vs low disease activity), or phase of therapy (induction vs maintenance). For example, in the study by Adedokun et al,40 the proposed threshold for infliximab was 22.0 mg/mL (week 6), which was associated with clinical response at week 8 compared with thresholds of 5.1 mg/mL and 3.7 mg/mL, which were associated with clinical response at week 14 and week 30, respectively. Therefore, provider and patient target thresholds can vary, depending on what scenario TDM is performed in, and a therapeutic range is more likely rather than a single fixed threshold. For example, when TDM is performed in the reactive setting, given limited treatment options, the primary focus may be to identify above what trough concentration threshold is a patient unlikely to respond to further dose escalation; alternatively, in a proactive setting, the primary focus of TDM may be to identify whether a patient is below a certain threshold at which risk of failure of therapy is high. We did not identify any RCTs or comparative studies comparing different target trough concentration thresholds of anti-TNF agents. We sought to evaluate the relative proportion of patients who were or were not in remission, above and below a target trough concentration. These data were derived largely from cross-sectional studies of maintenance therapy, wherein studies reported the sensitivity and specificity of one or more different thresholds in predicting the presence of clinical remission (or response). Most often, these data were reported as the value corresponding to the maximum area under the receiver-operator curve. To compare performance of different thresholds, we sought unpublished data from authors on performance of prespecified commonly used thresholds. We could report such data from infliximab, adalimumab, certolizumab pegol, and golimumab; there were very limited data on relative performance of different maintenance troughs for vedolizumab or ustekinumab.

Infliximab Trough Concentration Thresholds Data on relative performance of different infliximab trough concentration thresholds was derived from 6 studies,

-,

No.

-

which either reported or shared unpublished data for multiple predefined trough thresholds (Table 4).40,41,46
49 On analysis of different thresholds, proportion of patients not in remission progressively decreased from 25% when using an infliximab threshold of 1 mg/mL, to 15% with an infliximab trough concentration of 3 mg/mL, 8% with an infliximab trough concentration of 5 mg/mL and 4% with an infliximab trough concentration of 7 mg/mL or 10 mg/ mL (P < .001). Therefore, with increasing trough concentration thresholds of 7 mg/mL or 10 mg/mL, the proportion of infliximab-treated patients not in remission was not significantly different. In contrast, we observed that at higher infliximab trough concentration thresholds, a significant proportion of patients even below that threshold were in remission. Interestingly, when comparing the proportion of patients not in remission above a certain threshold for UC and CD separately, higher numbers of patients with UC were unlikely to be in remission at infliximab trough concentrations 3 mg/mL compared with patients with CD (Supplementary Table 3). However, only a small proportion of patients are not in clinical remission with an infliximab trough concentration 7 mg/mL, regardless of disease type. Therefore, at an infliximab trough concentration of 5 mg/mL during maintenance therapy, a small proportion of patients may not be in remission. Based on indirect evidence, there may be benefit to targeting higher trough concentrations in patients who are not responding at lower trough concentrations, although there may be only marginal benefit for targeting infliximab trough concentrations of 7 mg/mL for achieving clinical remission. Although a formal GRADE-based analysis of this evidence was not conducted, the included studies were at high risk of bias due to selective inclusion of patients. Inconsistency was observed across studies in trough thresholds and, therefore, we relied on unpublished data across prespecified thresholds. Also, due to paucity of direct evidence, indirect evidence from crosssectional studies is being used to extrapolate to potential benefits of escalation of therapy.

Adalimumab Trough Concentration Thresholds Data on relative performance of different adalimumab trough concentration thresholds was derived from 6 studies that either reported or shared unpublished data for multiple thresholds (Table 5).25,50
54 Of note, if studies did not report outcomes at multiple trough concentration, we used a cutoff for which data were reported for pooling. On analysis of different thresholds, proportion of patients not in remission progressively decreased from 17% when using an adalimumab threshold 5.0 ± 1 mg/mL, to 10% with an adalimumab trough concentration of 7.5 ± 1 mg/ mL. In contrast, we observed that at higher adalimumab trough concentration thresholds, a significant proportion of patients even below that threshold were in remission. There were limited data on analyzing differences in thresholds for UC and CD. Therefore, at an adalimumab trough concentration of 7.5 mg/mL during maintenance therapy, a small proportion of patients may not be in

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200

-

AGA Review of Therapeutic Drug Monitoring in IBD

11

Table 4.Distribution of Patients in Terms of Clinical Status, at Different Infliximab Cutoffs in Cross-Sectional or Cohort Studies Trough at or above cutoff, % (95% CI) Infliximab cutoff, mg/mL 1 3 5 7 10

Studies, n/ patients, n 640,41,46–49/929 6,40,41,46–49/929 640,41,46–49/929 540,46–49/726 440,41,47,48/772

Patients in remission 49.2 36.7 27.3 21.4 17.2

(38.8
59.6) (28.6
45.7) (20.9
34.9) (15.1
29.5) (11.8
24.6)

Patients not in remissiona 24.8 14.7 7.9 3.9 4.3

(17.7
33.7) (9.7
21.7) (4.6
13.2) (1.5
9.5) (1.8
9.6)

Trough below cutoff, % (95% CI) Patients in remission 9.5 20.7 29.7 36.0 43.0

(6.0
14.6) (14.0
29.5) (19.9
41.8) (24.3
49.6) (27.6
59.8)

Patients not in remission 15.5 25.9 32.8 36.4 35.7

(10.9
21.7) (20.8
31.7) (25.8
40.7) (25.9
48.3) (27.1
45.4)

NOTE. The proportion of patients not in remission above infliximab trough concentration thresholds decreases with increasing trough concentration. a P value for trend, for this group (most relevant) <.001.

remission. Based on indirect evidence, there may be benefit to targeting higher trough concentrations in patients who are not responding at lower trough concentrations.

Certolizumab Pegol Trough Concentration Thresholds For certolizumab pegol maintenance therapy, we included data from a large exposure
response pooled analysis, including data from 9 trials encompassing 2157 patients (Table 6).55 When focusing on the quadrant with certolizumab pegol trough concentration greater than or equal to the threshold and proportion of patients not in remission, a similar trend was apparent as for infliximab and adalimumab. Of the patients with a certolizumab pegol trough concentration of 10 mg/mL, 42% are not in remission compared with 32% with a certolizumab pegol trough concentration of 15 mg/mL and 26% with a certolizumab pegol trough concentration of 20 mg/mL. Therefore, targeting a certolizumab pegol trough concentration of 20 mg/mL can lead to an increased proportion of patients in clinical remission, with some patients benefiting from higher concentrations. In comparison to infliximab and adalimumab, certolizumab pegol concentration thresholds were considerably higher and so was the proportion of patients not in remission at those thresholds.

Golimumab Trough Concentration Thresholds Exposure
response data on golimumab is sparse compared with other anti-TNF agents. In a post-hoc analysis of the PURSUIT (Program of Ulcerative Colitis Research Studies Utilizing an Investigational Treatment) trials, Adedokun et al45 identified a golimumab concentration of 2.5 mg/mL at week 6 during induction and 1.4 mg/mL at week 44 (steady-state trough concentration) during the maintenance phase to be minimal thresholds to achieve clinically important outcomes (sensitivity 59.0%, specificity 67.2% and sensitivity 54.1%, specificity 77.6%, respectively). In a prospective observational study of 21 patients with UC, Detrez et al56 identified a golimumab threshold concentration of 2.6 mg/mL at week 6 to be associated with a clinical improvement at week 14 (90% specificity, 56% sensitivity, area under receiver-operator characteristic curve, 0.79; P ¼ .034). Discussion. In summary, when using reactive TDM during maintenance therapy, a trough concentration threshold of 5 mg/mL for infliximab, 7.5 mg/mL for adalimumab, and 20 mg/mL for certolizumab pegol, may identify most patients who are unlikely to respond to escalation of therapy. However, a small proportion of patients with ongoing inflammation despite adequate trough concentrations may be able to achieve clinical remission by targeting higher drug trough concentrations. In addition, these thresholds may be different between UC and CD, and patients with UC may require higher target trough

Table 5.Distribution of Patients in Terms of Clinical Status, at Different Adalimumab Cutoffs in Cross-Sectional or Cohort Studies Trough at or above cutoff, % (95% CI)

Trough below cutoff, % (95% CI)

Adalimumab cutoff, mg/mL

Studies, n/ patients, n

Patients in remission

Patients not in remissiona

Patients in remission

Patients not in remission

5.0 ± 1 7.5 ± 1

425,50
52/232 352–54/131

39.3 (6.3
54.0) 46.8 (34.8
59.2)

17.0 (12.7
22.5) 10.3 (6.1
16.9)

11.3 (7.7
16.2) 21.4 (15.2
29.3)

34.6 (25.2
45.3) 23.3 (16.8
31.4)

NOTE. The proportion of patients not in remission above adalimumab trough concentration thresholds decreases with increasing trough concentration. a P value for trend, for this group (most relevant) <.001.

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

AGA SECTION

1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260

2017

1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320

12

AGA SECTION

1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380

Casteele et al

Gastroenterology Vol.

-,

No.

-

Table 6.Distribution of Patients in Terms of Clinical Status, at Different Certolizumab Pegol Cutoffs in a Pooled Analysis of Individual Participant Data of 9 Trials (2157 Patients) of Certolizumab Pegol in Crohn’s Disease Trough at or above cutoff, % (n) Certolizumab pegol cutoff, mg/mL 10 15 20

Trough below cutoff, % (n)

Patients in remission

Patients not in remissiona

Patients in remission

Patients not in remission

21.3 (231) 16.6 (180) 12.9 (140)

41.6 (450) 32.2 (348) 25.5 (276)

11.6 (125) 16.3 (176) 20.0 (216)

25.5 (276) 34.9 (378) 41.6 (450)

NOTE. The proportion of patients not in remission above certolizumab pegol trough concentration thresholds decreases with increasing trough concentration. a P value for trend, for this group (most relevant) <.001.

concentrations compared with CD. These target trough thresholds may be higher if the treatment target is to achieve endoscopic remission, rather than clinical response/ remission. Although there are emerging data on optimal golimumab trough concentrations, the data are too sparse to allow for comprehensive assessment of different reported thresholds.

Thresholds for Anti-Drug Antibodies to AntiLTumor Necrosis Factor Agents In contrast to trough concentrations, quantitative results of ADAbs from different assays cannot be compared, and are typically interpreted qualitatively in the appropriate clinical context. Some studies have shown that patients with lowlevel ADAbs with detectable but subtherapeutic trough might still respond to escalation compared with patients with high-level ADAbs.57–59 In a cohort study of 125 patients with CD by Baert et al,60 an ADAb concentration of >8 mg/mL was considered “high” and correlated with a shorter duration of response in between infusions and a higher risk of infusion reactions (RR, 2.40; 95% CI, 1.65
3.66; P < .001). In a retrospective cohort study, it was observed that of the patients developing ADAbs to infliximab, 15 of 53 (28%) had transient ADAbs, and 38 of 53 (72%) had sustained ADAbs.57 A similar proportion of patients lost response to therapy during follow-up in both groups, 13 of 15 (87%) and 30 of 38 (79%) patients, respectively. However, of those patients undergoing empiric infliximab dose escalation at time of loss of response, 69% of patients with transient ADAbs recaptured response compared with 16% of patients with sustained ADAbs (P ¼ .0028). A concentration of ADAbs to infliximab of >9.1 U/mL was associated with an unsuccessful intervention (likelihood ratio, 3.6; specificity 82%, sensitivity 65%, area under the curve, 0.73; P ¼ .003). Of those patients with transient ADAbs, 2 of 15 (13%) had to discontinue infliximab therapy because of persistent loss of response or hypersensitivity reactions compared with 26 of 38 (68%) with sustained ADAbs (RR, 5.1; 95% CI, 1.4
19.0; P < .001). In addition, in the retrospective study of Yanai et al,24 a threshold for ADAbs to infliximab of >9 mg/mL equivalents and for ADAbs to adalimumab of

>4 mg/mL was proposed that identified patients with high ADAbs that did not respond to empiric dose escalation with 90% specificity.

Primer on Therapeutic Drug Monitoring for Thiopurines Thiopurine metabolism is complex and involves multiple conversion steps. Azathioprine (AZA) is a prodrug and enzymatic and nonenzymatic conjugation via glutathione is necessary to convert it into 6-mercaptopurine (6-MP).61,62 Several enzymes are involved in modification of 6-MP into active (6-TGN) and inactive (6-MMP, 6methylmercaptopurine ribonucleotides, thiouric acid) metabolites. The immunosuppressive effect of thiopurines is achieved by incorporation of 6-TGN into nucleic acids instead of guanine nucleotides, which inhibits purine and protein synthesis in lymphocytes. TPMT is one of the key enzymes in thiopurine metabolism. In patients with low or absent TPMT enzyme activity, thiopurine exposure can lead to severe bone marrow toxicity due to the excess production of drug-derived TGN metabolites. TPMT enzyme activity is controlled by >25 genetic polymorphisms, but the most common 4 alleles are found in 80%
95% of tested persons.63 Approximately 4%
11% of the population are heterozygous for a variant TPMT allele and have intermediate enzyme activity, and around 0.3% are homozygous and have very low or absent enzyme activity.63 TPMT activity has been associated with risk of myelosuppression, but not other adverse events, in thiopurine-treated patients. In a recent systematic review of 54 observational studies and 1 randomized trial of TPMT status in adults and children with inflammatory diseases, the odds of thiopurine-induced leukopenia for heterozygous and homozygous genotypes was 4.3 (95% CI, 2.7
6.9) and 20.8 (95% CI, 3.4
126.9), respectively.63 Most events of leukopenia occur within 8 weeks of starting therapy, however, leukopenia can be observed after many months of thiopurine therapy.64,65 Based on risk of myelosuppression in patients with low or absent TPMT enzyme activity, the Food and Drug Administration recommends genotype or phenotype testing before high-dose chemotherapy with thiopurines in pediatric patients.66

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440

1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500

2017

AGA Review of Therapeutic Drug Monitoring in IBD

The therapeutic effect of thiopurines has been associated with the level of active 6-TGN metabolites, and hepatotoxicity has been associated with higher levels of the inactive metabolites 6-MMP and 6-methylmercaptopurine ribonucleotides.5,67,68 TDM of thiopurines may offer the possibility to improve patient outcomes by identifying symptomatic patients with low 6-TGN levels. Additionally, by measuring 6-MMP levels, a subgroup of patients can be identified who preferentially convert 6-MP to 6-MMP and often do not achieve sufficient 6-TGN levels. This group of patients, often described as “shunters,” may be susceptible to hepatotoxicity because thiopurine dose escalation leads to 6-MMP accumulation.

Results Question 3: In Patients With Inflammatory Bowel Disease Being Started on Thiopurines, Is Routine Thiopurine Methyltransferase Measurement (to Guide Dosing) Superior to No Thiopurine Methyltransferase Measurement (With Empiric Weight-Based Dosing of Thiopurines)? Key message. The benefit of routine TPMT testing (to guide initial dosing of thiopurines) over no TPMT testing (with empiric weight-based dosing) is uncertain for most patients, but may avoid serious harm in a small fraction of patients (low-quality evidence). Effect estimates. Three RCTs comparing a strategy of TPMT testing (to guide thiopurine dosing) vs no TPMT testing (with empiric weight-based thiopurine dosing) were identified.69–71 In the intervention arm, thiopurine dose recommendations were based on TPMT genotype (2 trials) or enzyme activity (1 trial), such that patients with wildtype genotype (or normal enzyme activity) were advised to start 2
2.5 mg/kg/d AZA (or 1
1.5 mg/kg/d 6-MP); patients heterozygous for TPMT genotype (or with intermediate enzyme activity) were advised to start at lower doses (50% dose reduction), and patients homozygous for TPMT genotype (or with low/absent enzyme activity) were either advised to avoid thiopurines or started at 0
10% of standard dose. In contrast, patients in the control arms of these trials were advised to start standard weight-based dose of thiopurines upfront (2 trials) or with gradual titration (1 trial); regardless of intervention, biochemical parameters, including complete blood count and liver enzymes, were closely monitored in both treatment arms. Among 1145 patients included in trials, 1032 (90.1%) had wild-type genotype (or normal enzyme activity), 111 patients (9.7%) were heterozygous, and 2 patients (0.17%) were homozygous. None of the trials reported rates of serious adverse events; hematologic adverse events, primarily neutropenia, and treatment discontinuation, were considered surrogates for outcome of interest. Observed rates of hematologic adverse events (TPMT testing vs no TPMT testing: 32 of 582 vs 32 of 552; RR, 0.94; 95% CI, 0.59
1.50) and treatment discontinuation (225 of 582 vs 195 of 552; RR, 1.09; 95% CI, 0.94
1.27) were not significantly different between the treatment groups (Table 7);

13

rates of other adverse events like pancreatitis or hepatotoxicity were also not significantly different. Six patients died in 2 trials, including 2 patients with sepsis-related deaths (1 in each arm); the other 4 deaths were deemed unrelated to AZA (RR, 0.48; 95% CI, 0.09
2.62). No significant difference was observed in rates of clinical remission between treatment arms, in a subset of patients with baseline active symptoms (RR, 1.03; 95% CI, 0.84
1.27), or in mean disease activity scores.

Quality of evidence Overall, the trials were deemed to be at low risk of bias; though physicians were not aware of TPMT genotype status, true blinding was not feasible because treatment advice on choice of thiopurine starting dose allowed inference of TPMT genotype in the intervention arm. While TPMT enzyme activity is more commonly available and used in clinical practice in the United States, TPMT genotype testing correlates closely with TPMT enzyme activity levels and, therefore, interventions were deemed comparable across trials63; in addition, although trials presented events based on genotype, results for TPMT enzyme activity were reported to be similar. Because information on serious adverse events was not reported directly, hematologic adverse events and treatment discontinuation were deemed surrogate outcomes and evidence was rated down for indirectness. Evidence was also rated down for serious imprecision due to a wide CI crossing unity. The overall quality of evidence was rated as low, that is, there was low confidence in the effect estimates supporting the use of routine TPMT testing (to guide thiopurine dosing) over no TPMT testing (with empiric weight-based dosing) in reducing risk of serious adverse events. Potential harms of intervention. There is minimal risk to performing TPMT genotype or enzyme activity as a blood test. While testing for TPMT may potentially delay starting therapy for approximately 1
2 weeks (while awaiting test results), it is likely inconsequential given the slow onset of action of this medication. TPMT testing neither increases nor alleviates the burden of routine laboratory monitoring needed for patients treated with thiopurines because laboratory monitoring is obligatory for all thiopurine-treated patients, regardless of TPMT status. On the other hand, for a very small fraction of patients who have very low or absent enzyme activity, no TPMT testing with empiric weight-based dosing of thiopurines can conceivably result in early and severe leukopenia and associated complications before the first scheduled laboratory monitoring. Discussion. TPMT activity can be evaluated by either genotyping or directly measuring the TPMT activity in red blood cells by radiochemical or chromatographic techniques in vitro (phenotyping).62 For this review, we did not evaluate the different methodologies used to determine TPMT activity. There is an ongoing debate about the advantages and disadvantages of a genotyping vs phenotyping approach, which is discussed in depth in other publications.63,72 It is important to note that blood transfusions in a

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

AGA SECTION

-

1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560

1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 AGA SECTION

14 Casteele et al Table 7.In Patients With Inflammatory Bowel Disease Being Started on Thiopurines, Is Routine Thiopurine Methyltransferase Enzyme Activity Or Genotype Measurement (to Guide Dosing) Superior to No Thiopurine Methyltransferase Measurement (With Empiric Weight-Based Dosing of Thiopurines)?69–71 PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

Summary of findings

Quality assessment Study event rate,

Anticipated absolute effects

n (%) No. of participants (studies), follow-up Discontinuation of therapy (critical outcome) 1134 (3 RCTs)

Risk of bias

Inconsistency Indirectness Imprecision

Publication Overall quality bias of evidence

With no TPMT testing

With TPMT testing

Risk difference Relative effect, Risk with no with TPMT testing RR (95% CI) TPMT testing

Not serious

Not serious

Seriousa

Seriousb

None

44 Low

195/552 (35.3)

225/582 (38.7)

1.09 (0.94
1.27)

353 per 1000

32 more per 1000 (21 fewer to 95 more)

Hematologic adverse events (critical outcome) 1134 (3 RCTs) Not serious

Not serious

Seriousa

Seriousb

None

44 Low

32/552 (5.8)

32/582 (5.5)

0.94 (0.59
1.50)

58 per 1000

3 fewer per 1000 (24 fewer to 29 more)

Gastroenterology Vol.

RR, risk ratio. a Surrogate outcomes for critical outcome of interest (serious adverse events). b Wide CIs crossing unity, and unable to exclude high rate of benefit or harm.

-,

No. -

1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680

1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 Q8 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740

2017

AGA Review of Therapeutic Drug Monitoring in IBD

3-month time interval before TPMT phenotyping might affect the analyses and yield unreliable results. Despite observational studies reporting significantly higher risk of myelosuppression in a subset of patients treated with thiopurines, the quality of evidence supporting routine TPMT testing to guide thiopurine dosing was deemed low based on RCTs. This can be attributed to 2 factors. First, the frequency of heterozygous (9.7%) and homozygous (0.17%) genotype in included trials is low, comparable to the general population. Coenen et al69 performed a post-hoc subgroup analysis of their large Dutch trial and demonstrated a significant benefit of TPMT pretesting for the heterozygous carriers with 89% risk reduction (RR, 0.11; 95% CI, 0.01
0.85) for hematologic adverse events. Based on these data, they calculated that a prospective trial would need more than 42,000 participants to show a significant benefit for the whole patient population. Second, a significant proportion of thiopurine-related hematologic adverse events occur despite normal TPMT status, emphasizing that TPMT testing cannot substitute for routine complete blood count (CBC) monitoring in patients receiving thiopurines.64,73 The package insert for AZA recommends monitoring of CBC weekly during the first month, twice monthly for the second and third months of treatment, then monthly or more frequently if dosage alterations or other therapy changes are necessary. However, patient and physician compliance with regular CBC monitoring is suboptimal; in a cohort study, the median completion of CBC monitoring in the first 8 weeks and between weeks 9 and 26 of therapy was 3 CBCs (interquartile range, 1
4) and 1 CBCs (interquartile range, 1
2), respectively.65 Although routine laboratory monitoring is necessary with thiopurine use in IBD patients, the optimal frequency of monitoring is uncertain.74 Given the low prevalence of heterozygous and homozygous TPMT genotypes, the magnitude of population-level benefit of routine TPMT testing is likely to be very small. However, there is minimal risk to performing this test, and it neither increases nor alleviates the potential burden of routine laboratory monitoring needed for patients treated with thiopurines. Contemporary cost-effectiveness analyses of routine TPMT testing are lacking. In a cost-effectiveness analysis based on the UK TARGET trial, there were marginal cost-savings, but these results may not apply to the United States because costs of TPMT testing vary significantly across health care systems.75

Question 4. In Patients With Active Inflammatory Bowel Disease Treated With Thiopurines or With Side Effects Thought to be Due to Thiopurine Toxicity, Is Reactive Therapeutic Drug Monitoring to Guide Treatment Changes Superior to No Therapeutic Drug Monitoring With Empiric Treatment Changes? Key message. In patients with IBD treated with thiopurines with active IBD-related symptoms or side effects thought to be due to thiopurine toxicity, there may be a

15

benefit of reactive TDM to guide treatment changes over empiric treatment changes (very-low-quality evidence). Effect estimates. We did not identify any randomized trials or prospective comparative studies in thiopurinetreated IBD patients comparing reactive TDM to guide treatment changes vs empiric treatment changes. One retrospective single-group observational study reported outcomes in thiopurine-treated symptomatic IBD patients when treatment changes were concordant vs discordant with suggestions based on a reactive TDM management algorithm (Supplementary Table 4).76 Based on this algorithm, 25 of 63 patients were considered refractory to thiopurine therapy despite appropriate dosing (6-TGN, 230
450 pmol/8  108 RBCs; normal or high 6-MMP), 18 were underdosed (6-TGN <230 pmol/8  108 RBCs, normal 6-MMP), 6 were noncompliant, 8 were overdosed (6-TGN >450 pmol/8  108 RBCs), and 6 patients were considered shunters (low 6-TGN, high 6-MMP, with 6-MMP/6TGN ratio >11). Among patients deemed refractory to thiopurine therapy based on TDM algorithm, 12 of 12 patients (100%) in whom treatment changes were concordant with suggestions from algorithm (switching to alternative therapy, without further dose escalation) improved, and only 1 of 13 patients (7.7%) improved when treatment changes were discordant from suggested algorithm; of 4 patients in whom thiopurine dose was increased despite “therapeutic” levels, 3 failed to respond and 1 developed toxicity. Among underdosed patients, 13 of 15 patients (86.7%) with algorithm-concordant dose escalation achieved clinical response; treatment discordance in the form of failure to increase dose resulted in failure to achieve clinical response (1 of 1), whereas addition of alternative therapy like steroids or anti-TNF could achieve response in 2 patients. Of 7 patients deemed to be failing thiopurines despite supratherapeutic 6-TGN levels, algorithm-concordant switching to alternative therapy with or without decrease in thiopurine dose resulted in clinical response in 5 of 5 patients (100%), and 2 patients in whom thiopurine dose was decreased or unchanged without any associated treatment changes failed to respond. Among 6-MMP shunters, a decrease in thiopurine dose with either addition of allopurinol or other alternative therapies, concordant with management algorithm resulted in clinical response in 4 of 4 patients (100%), and only a decrease in thiopurine dose did not result in clinical improvement in 2 of 2 patients (although abnormal liver enzymes resolved). Based on this single study, considerable differences were observed in patient outcomes if treatment decisions were concordant with the proposed metabolite-based algorithm compared with empiric treatment changes, a proportion of which may be discordant with algorithm. Overall, algorithm-concordant treatment changes were significantly more likely to result in clinical response in symptomatic patients compared with algorithm-discordant treatment changes (36 of 42 vs 3 of 18; RR, 5.15; 95% CI, 1.82
14.56) (Table 8). Quality of evidence. Due to the observational nature of the study, with inherent risk of bias of unmeasured confounding, along with indirectness of comparisons (instead of directly comparing reactive TDM vs empiric

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

AGA SECTION

-

1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800

16

AGA SECTION

1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860

Casteele et al

Gastroenterology Vol.

treatment changes, the study compared results for algorithm concordant vs discordant treatment decisions) and imprecision due to small study size, overall quality of evidence was rated as very low quality, that is, there was very low confidence in the effect estimates supporting the use of reactive TDM (to guide thiopurine dosing) over empiric treatment changes in thiopurine-treated patients with inadequate clinical response or side effects. Potential harms of intervention. TDM is performed through a blood test in patients on a stable dose of thiopurines and can cause mild inconvenience. The performance of therapeutic 6-TGN cutoffs is suboptimal in discriminating those who may or may not respond to specific treatment changes; in addition, target 6-TGN cutoffs when thiopurines are used as concomitant therapy with biologics to decrease risk of immunogenicity are poorly defined. Incidental findings during thiopurine metabolite testing can potentially result in unwarranted treatment changes. With the availability of several newer and more effective therapeutic agents over the last decade, attempts at close metabolite monitoring and serial dose adjustments can potentially delay more effective therapy in a subset of underdosed patients who may be inherently thiopurine-resistant. In contrast, empiric treatment changes, such as escalation of therapy in patients with suboptimal response, may result in excessively high 6-TGN level, which increases risk of leukopenia, or excessively high 6-MMP levels due to shunting, which increases risk of hepatotoxicity. Inappropriate treatment changes can also potentially delay use of more effective therapy. Discussion. Reactive drug monitoring in patients with active symptoms or drug-related side effects assumes that adaption of a drug level to a suggested drug concentrations results in a significant improvement or disappearance of the side effect. However, the performance of the most commonly used 6-TGN cutoff (>230 pmol/8  108 RBCs) is suboptimal (see discussion of routine thiopurine TDM). Additionally, when applying these cutoffs to reactive TDM, there is potential risk of bias because most studies evaluating this cutoff were cross-sectional and lacked standardization of nucleotide assays. There is paucity of prospective studies evaluating whether achieving proposed target 6TGN levels would be associated with higher rates of clinical remission. Available evidence does not inform whether these proposed cutoffs might be different in patients with CD and UC. It is unclear whether optimal cutoffs of 6-TGN may be different in patients on thiopurine monotherapy (where the goal is controlling disease activity) and patients in whom thiopurines are used in conjunction with anti-TNF agents or other biologic agents (where the primary goal may be to decrease immunogenicity). In a cross-sectional study, Yarur et al77 observed a moderate correlation between 6TGN levels and anti-TNF levels, and a 6-TGN cutoff >125 pmol/8  108 RBCs was associated with high anti-TNF and mucosal healing compared with patients with 6-TGN <125 pmol/8  108 RBCs. While anti-TNF level was numerically higher in patients with 6-TGN >252 pmol/8  108 RBCs compared with patients with 6-TGN 125
176 pmol/8  108 RBCs (17.8 mg/mL vs 13.4 mg/mL), this difference was

-,

No.

-

not statistically significant (P ¼ .12). However, in another cross-sectional study, there was no correlation between 6-TGN levels and anti-TNF levels, with limited ability of 6TGN >125 pmol/8  108 RBCs to discriminate achievement of adequate vs inadequate trough levels (unpublished data, personal communication with Viraj Kariyawasam, Mark Ward, and Peter Irving). Two studies laid the foundation for reactive TDM for thiopurines showing that an increase of thiopurine dosing in a subgroup of patients with active IBD was associated with a significant increase of 6-TGN concentrations and clinical remission.5,78 However, in both studies, the dose adaptation in the individual patient was not performed based on trough levels, but solely on clinical judgment, and specific cutoffs were derived retroactively. Many patients in both studies were not dosed according to the recommended weightbased algorithm (2
2.5 mg/kg body weight for AZA and 1
1.5 mg/kg body weight for 6-MP). Dubinsky et al5 also observed that in some patients, thiopurine dose increase resulted mainly in a shunting of thiopurine metabolism toward 6-MMP rather than 6-TGN, and these patients failed to achieve remission but rather developed hepatotoxicity in the presence of high 6-MMP levels. Four noncomparative, retrospective studies have reported the efficacy of reactive TDM, which had been initiated in thiopurine-treated patients because of active IBD despite thiopurine therapy, suspicion of lack of adherence, or adverse events thought to be related to thiopurine therapy.79–82 The end points for the evaluation of a successful reactive TDM in the individual patients in these studies varied and included change in disease management (adaption of drug dosing, switch to another drug, or surgery), improved clinical outcome or remission based on physician assessment or resolution of side effects thought to be thiopurine related; however, none of these studies reported outcomes with empiric treatment changes and did not contribute to summary estimates. None of the studies used established diseases activity scores, such as the Mayo or the Harvey Bradshaw score, or assessed for mucosal healing. The outcomes are depicted in Supplementary Table 5. Of note, noncompliance occurred in these retrospective studies in 6%
11%.76,82 Overall, the data suggest that in more than half of the patients, adoption of the therapeutic approach based on trough levels was successful in achieving a better outcome, defined as either response or remission, and that in more than one third of patients, a resolution of the thiopurine-associated side effects could be accomplished.

Question 5. In Patients With Inflammatory Bowel Disease Treated With Thiopurines, Is Routine Therapeutic Drug Monitoring to Guide Thiopurine Dosing Superior to Empiric Weight-Based Dosing? Key message. The benefit of routine TDM to guide thiopurine dosing over empiric weight-based dosing is uncertain (very-low-quality evidence). Effect estimates. Two small, randomized studies, one double-blind RCT in the United States, another open-label

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920

1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 -

2017

Summary of findings

Quality assessment Study event rates, n (%) No. of participants (studies), follow-up

Risk of bias

Anticipated absolute effects

Risk with With reactive empiric With empiric TDM-guided treatment Relative effect, treatment Publication Overall quality treatment changes changes changes RR (95% CI) Inconsistency Indirectness Imprecision bias of evidence

Clinical response (critical outcome) 60 (1 study) Not serious Not serious

Seriousa

Seriousb

ND

4 Very low

3/18 (16.7)

36/42 (85.7)

5.15 (1.82
14.56)

167 per 1,000

Risk difference with reactive TDM-guided treatment changes

692 more per 1000 (137 more to 2260 more)

ND, not detected; RR, risk ratio. a Outcomes reported based on algorithm-concordant care vs algorithm discordant care, and not empiric treatment changes; this may potentially overestimate benefit of intervention. b Small study size with low event rate (optimal information size not met).

AGA Review of Therapeutic Drug Monitoring in IBD

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

Table 8.In Patients With Active Inflammatory Bowel Disease Treated With Thiopurines or With Side Effects Thought to Be Due to Thiopurine Toxicity, Is Reactive Therapeutic Drug Monitoring to Guide Treatment Changes Superior to no Therapeutic Drug Monitoring With Empiric Treatment Changes?76

17

AGA SECTION

1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040

18

AGA SECTION

2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 Q9 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100

Casteele et al

Gastroenterology Vol.

randomized trial conducted in Germany, investigated scheduled thiopurine metabolite testing with successive adaptation of AZA therapy to a target 6-TGN concentration of 250
400 pmol/8  108 RBCs vs standard AZA weightbased dosing (2.5 mg/kg body weight).83,84 In the US trial, the intervention involved both TPMT testing to guide initial dosing, followed by prospective 6-TGN
guided dose adaptation compared with empiric weight-based dosing with gradual dose escalation if well tolerated (regardless of TPMT activity) in control arm.83 In contrast, in the German trial, all patients were started on TPMT-level
guided dosing initially, followed by randomization to 6-TGN
guided dose adaptation or continued weight-based dosing.84 On pooled analysis of both trials, at 16 weeks there was a numerically higher proportion of patients achieving clinical remission in patients who underwent routine TDM-guided dose adaptation compared with standard weight-based dosing (21 of 50 [42%] vs 18 of 47 [31.6%]), but the difference was not statistically significant (RR, 1.44; 95% CI, 0.59
3.52) (Table 9). The rate of serious adverse events (requiring discontinuation of therapy) was comparable between the 2 arms (TDM-guided dose adaptation vs empiric dosing: 16 of 50 [32.0%] vs 15 of 57 [26.3%]; RR, 1.20; 95% CI, 0.50
2.91). Quality of evidence. Both studies were terminated early due to slow recruitment and failure to meet prespecified enrollment targets, putting them at high risk of bias. Additionally, there was high attrition rate in both trials (33%
46%), although the analyses were conducted in intention-to-treat manner with worst-case scenario imputation. Due to the small number of events and wide CIs crossing unity for both critical outcomes, evidence was further rated down for very serious imprecision. The overall quality of evidence was rated as very low quality, that is, there was very low confidence in the effect estimates supporting the use of routine TDM-guided dose adaptation over weight-based thiopurine dosing for improving clinical outcomes. Potential harms of intervention. Routine monitoring of thiopurine metabolites is conducted through a blood test, which imposes minimal risk to the patients. However, with each dose adaptation, intensified laboratory monitoring is required for early identification of neutropenia or hepatotoxicity, which can be inconvenient and expensive. Given slow onset of action of this medication, adaptation is conducted slowly over weeks, which can potentially delay administration of alternative effective therapies in patients with inadequate response to thiopurines. Discussion. Based on 2 small RCTs at high risk of bias, there is uncertainty whether routine 6-TGN measurement and adaptive dosing are superior to standard weight-based dosing based on TPMT levels. The included trials focused on utility of trying to identify potentially underdosed patients (6-TGN <230 pmol/8  108 RBCs), and then trying to adapt thiopurine dosing to achieve 6-TGN level >230 pmol/8  108 RBCs. It is important to note that although this 6-TGN target >230 pmol/8  108 RBCs is often cited, the performance characteristics of this cutoff are suboptimal. In a

-,

No.

-

recent meta-analysis of 17 studies in 2052 patients with IBD (12 adult and 5 pediatric IBD cohorts), 15 were designed as cross-sectional studies, evaluating distribution of 6-TGN levels in patients in varying stages of disease activity; only 2 studies were prospective cohort studies evaluating the longitudinal impact of 6-TGN >230 pmol/8  108 RBCs on disease activity.68 In these studies, 53.6% of patients were in clinical remission at the time of 6-TGN measurement. The overall sensitivity and specificity of 6-TGN >230 pmol/8  108 RBCs in identifying patients in remission was 63% and 53%, respectively (area under receiver-operator characteristic curve, 0.63). Among patients with active disease, 66% had 6-TGN <230 pmol/8  108 RBCs and 34% had 6-TGN >230 pmol/8  108 RBCs; however, in patients in remission, 50% patients had 6-TGN >230 pmol/8  108 RBCs and 50% patients had 6-TGN <230 pmol/8  108 RBCs. Patients with 6-TGN >230 pmol/8  108 RBCs were 1.4 times more likely to be in clinical remission (RR, 1.36; 95% CI, 1.16
1.61) compared with patients with 6-TGN <230 pmol/8  108 RBCs, with considerable heterogeneity. It is frequently perceived that in patients with low 6-TGN levels, with increase in dose, a proportion of these patients might achieve higher 6-TGN levels. However, in both randomized studies, only a small proportion of patients in the intervention arm actually achieved the target 6-TGN, despite attempts at dose adaptation. In the US trial, despite significantly higher azathioprine dose at week 16 in the intervention arm (median dose of AZA, 3.4 vs 2.3 mg/kg/d in the individualized and weight-based dosing arms, respectively; P < .01), there was no significant difference in 6-TGN levels of 6-TGN concentrations (median, 197 pmol/8  108 RBCs; range, 81
286 pmol/8  108 RBCs vs median, 150 pmol/ 8  108 RBCs; range, 81
297 pmol/8  108 RBCs) in the individualized and weight-based dosing arms, respectively (P ¼ NS).83 This may be related to a change in metabolism of AZA away from active 6-TGN metabolites toward inactive metabolites, such as 6-MMP with dose increase.5,85 In both trials, due to safety concerns only a limited increase in AZA dose was allowed as a strategy to achieve the target 6-TGN level (maximal dose limit set at 3 mg/kg AZA in the German study and 4 mg/kg in the US study); interventions, such as adding allopurinol to modify thiopurine metabolism preferentially toward 6-TGN, were not allowed. There were several limitations of trial design, which failed to fully elucidate the potential benefit of routine thiopurine metabolite monitoring. Besides identifying underdosed patients (low 6-TGN level, normal 6-MMP levels), knowledge of thiopurine metabolites can facilitate the identification of the following subgroups: noncompliant patients (absent/very low 6-TGN and absent/very low 6MMP), preferential 6-MMP producers/shunters (low 6TGN levels, with high 6-MMP levels), treatment refractory patients (therapeutic 6-TGN with normal 6-MMP) or patients with high drug levels at risk of toxicity (high 6-TGN and high 6-MMP levels). While in clinical practice, it is conceivable that practitioners may adopt different strategies beyond AZA dose, depending on pattern of thiopurine metabolites (eg, adding allopurinol in shunters,86,87 switching out of treatment class if active symptoms occur despite

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160

2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 -

2017

Table 9.In Patients With Inflammatory Bowel Disease Treated With Thiopurines, Is Routine Therapeutic Drug Monitoring (Thiopurine Metabolite Monitoring) to Guide Thiopurine Dosing Superior to Empiric Weight-Based Dosing?83,84

Anticipated absolute effects

Study event rates, n (%) No. of participants (studies), follow-up

Risk of bias

Clinical remission (critical outcome) 107 (2 RCTs) Not serious

Discontinuation of therapy (due to drug-related adverse events) (critical outcome) 107 (2 RCTs) Not serious

Risk difference Risk with with thiopurine With With thiopurine metaboliteweightmetaboliteOverall weightguided guided Relative effect, based Publication quality of based dosing dosing dosing RR (95% CI) Inconsistency Indirectness Imprecision bias evidence dosing

Seriousa

Seriousa

Not serious

Not serious

Very seriousb

Very seriousb

None

None

44 LOW

18/57 (31.6%)

(42.0)

44 LOW

15/57 (26.3)

(32.0)

21/50

1.44 (0.59
3.52)

316 per 1000

139 more per 1000 (129 fewer to 796 more)

16/50

1.20 (0.50
2.91)

263 per 1000

53 more per 1000 (132 fewer to 503 more)

RR, risk ratio. a2 I > 50%. b Wide CIs crossing 1, and low event rate.

AGA Review of Therapeutic Drug Monitoring in IBD

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

Summary of findings

Quality assessment

19

AGA SECTION

2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280

20

AGA SECTION

2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340

Casteele et al

Gastroenterology Vol.

therapeutic 6-TGN), these clinical trials did not allow for these potential therapeutic interventions and did not study the entire scope of possible interventions with routine knowledge of thiopurine metabolites. Additionally, included clinical trials were unable to address the potential benefit of routine 6-TGN monitoring in identifying and appropriately treating noncompliant patients, given low observed rates of noncompliance in clinical trials (Hawthorne effect). Finally, both of these trials were conducted almost a decade ago, and since the conduct of these trials, treatment paradigms have evolved, including consideration for early introduction of biologic therapies in the management of IBD patients, considering a “top-down” strategy over a “step-up” approach, which limits the applicability of these trials in the current scenario. Current evidence is very limited in terms of understanding the potential benefit of routine thiopurine metabolite testing over standard weight-based dosing in thiopurine-treated patients. There are very limited data on the cost-effectiveness of a strategy of routine thiopurine metabolite monitoring.

Knowledge Gaps and Future Directions We have identified the following key knowledge gaps in the role of TDM in IBD, particularly with biologic agents (anti-TNF and non-anti-TNF), which should be addressed in properly designed studies. Observational and comparative evidence is needed to define minimal effective exposure thresholds that are associated with clinically meaningful outcomes after induction and maintenance therapy; similarly, the maximum threshold concentration beyond which a ceiling effect is observed (ie, above which further attempts at increased trough concentrations is highly unlikely to be effective) needs to be identified, acknowledging that such thresholds may be different for different outcomes of interest (eg, clinical remission, endoscopic remission, fistula healing, management of CD after surgically induced remission, and left-sided UC vs pan-UC). Once thresholds are identified, randomized trials comparing the efficacy and safety of early optimized therapy based on TDM to target trough concentration(s) vs standard induction dosing should be evaluated. For patients on maintenance therapy, adequately powered and well-designed randomized studies (with appropriate trough concentration and ADAb threshold defined algorithms) are needed to compare reactive TDM vs routine proactive TDM vs empiric treatment changes, evaluating long-term patient-important outcomes. To ascertain the frequency of performing measurements in the setting of proactive TDM, adequately powered randomized studies comparing, for example, one-time proactive TDM vs annual proactive TDM vs more frequent TDM, are warranted. These trials may need to be performed in different disease phenotypes, as differences in exposure
response relationships (eg, CD vs UC) became apparent for infliximab. With the availability of different drug assays, inter-assay reliability assessments for both trough concentrations and ADAbs are urgently warranted to guarantee reproducible clinical decision making across the different diagnostic platforms. Besides trough concentrations, comprehensive analyses of

-,

No.

-

persistent vs transient ADAbs and low- and high-titer ADAbs need to be performed to optimally inform TDM algorithms. Once these thresholds are well-defined, then clinical trials comparing exposure-based dosing vs fixed dosing are required to facilitate optimization of biologic agents. There is a lack of data on the exposure
response relationship of biologics with a mechanism of action other than blockage of TNF, which commands in-depth studies linking exposure to target engagement and patientimportant outcomes, instead of direct extrapolation from anti-TNF agents. Current therapeutic approaches favor a combination therapy of a thiopurine with a biologic rather than monotherapy. Prospective evaluations regarding target 6-TGN concentrations to prevent antibody formation or to optimize clinical efficacy in patients on a concomitant biologic and a thiopurine are warranted.

Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at www.gastrojournal.org, and at http://dx.doi.org/10.1053/ j.gastro.2017.07.031.

References 1. Colombel JF, Narula N, Peyrin-Biroulet L. Management strategies to improve outcomes of patients with inflammatory bowel diseases. Gastroenterology 2017; 152:351–361. 2. Vande Casteele N, Feagan BG, Gils A, et al. Therapeutic drug monitoring in inflammatory bowel disease: current state and future perspectives. Curr Gastroenterol Rep 2014;16:378. 3. Yarur AJ, Rubin DT. Therapeutic drug monitoring of antitumor necrosis factor agents in patients with inflammatory bowel diseases. Inflamm Bowel Dis 2015;21:1709–1718. 4. Afif W, Loftus EV Jr, Faubion WA, et al. Clinical utility of measuring infliximab and human anti-chimeric antibody concentrations in patients with inflammatory bowel disease. Am J Gastroenterol 2010;105:1133–1139. 5. Dubinsky MC, Yang H, Hassard PV, et al. 6-MP metabolite profiles provide a biochemical explanation for 6-MP resistance in patients with inflammatory bowel disease. Gastroenterology 2002;122:904–915. 6. Goldberg R, Irving PM. Toxicity and response to thiopurines in patients with inflammatory bowel disease. Expert Rev Gastroenterol Hepatol 2015;9:891–900. 7. Sultan S, Falck-Ytter Y, Inadomi JM. The AGA institute process for developing clinical practice guidelines part one: grading the evidence. Clin Gastroenterol Hepatol 2013;11:329–332. 8. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177–188. 9. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–560. 10. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–634.

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400

2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460

2017

AGA Review of Therapeutic Drug Monitoring in IBD

11. Lee TW, Fedorak RN. Tumor necrosis factor-alpha monoclonal antibodies in the treatment of inflammatory bowel disease: clinical practice pharmacology. Gastroenterol Clin North Am 2010;39:543–557. 12. Schellekens H, Casadevall N. Immunogenicity of recombinant human proteins: causes and consequences. J Neurol 2004;251(Suppl 2):II4–II9. 13. Ordas I, Feagan BG, Sandborn WJ. Therapeutic drug monitoring of tumor necrosis factor antagonists in inflammatory bowel disease. Clin Gastroenterol Hepatol 2012;10:1079–1087. 14. Papamichael K, Gils A, Rutgeerts P, et al. Role for therapeutic drug monitoring during induction therapy with TNF antagonists in IBD: evolution in the definition and management of primary nonresponse. Inflamm Bowel Dis 2015;21:182–197. 15. Marini JC, Sendecki J, Cornillie F, et al. Comparisons of serum infliximab and antibodies-to-infliximab tests used in inflammatory bowel disease clinical trials of Remicade®. AAPS J 2017;19:161–171. 16. European Medicines Agency. Guideline on Bioanalytical Method Validation. London: European Medicine Agency, 2009. 17. US Food and Drug Administration; Guidance for Industry: Bioanalytical Method Validation. Silver Spring, MD: US Food and Drug Administration, 2013. 18. Steenholdt C, Ainsworth MA, Tovey M, et al. Comparison of techniques for monitoring infliximab and antibodies against infliximab in Crohn’s disease. Therapeutic Drug Monitor 2013;35:530–538. 19. Martin S, del Agua AR, Torres N, et al. Comparison study of two commercially available methods for the determination of golimumab and anti-golimumab antibody levels in patients with rheumatic diseases. Clin Chem Lab Med 2015;53:e297–e299. 20. Bodini G, Giannini EG, Furnari M, et al. Comparison of two different techniques to assess adalimumab trough levels in patients with Crohn’s disease. J Gastrointes Liver Dis 2015;24:451–456. 21. Vande Casteele N, Khanna R, Levesque BG, et al. The relationship between infliximab concentrations, antibodies to infliximab and disease activity in Crohn’s disease. Gut 2015;64:1539–1545. 22. Brandse JF, Van Den Brink GR, Wildenberg ME, et al. Loss of infliximab into feces is associated with lack of response to therapy in patients with severe ulcerative colitis. Gastroenterology 2015;149:350–355. 23. Paul S, Del Tedesco E, Marotte H, et al. Therapeutic drug monitoring of infliximab and mucosal healing in inflammatory bowel disease: a prospective study. Inflamm Bowel Dis 2013;19:2568–2576. 24. Yanai H, Lichtenstein L, Assa A, et al. Levels of drug and antidrug antibodies are associated with outcome of interventions after loss of response to infliximab or adalimumab. Clin Gastroenterol Hepatol 2015;13:522–530. 25. Roblin X, Marotte H, Rinaudo M, et al. Association between pharmacokinetics of adalimumab and mucosal healing in patients with inflammatory bowel diseases. Clin Gastroenterol Hepatol 2014;12:80–84.

21

26. Steenholdt C, Brynskov J, Thomsen O, et al. Individualised therapy is more cost-effective than dose intensification in patients with Crohn’s disease who lose response to anti-TNF treatment: a randomised, controlled trial. Gut 2014;63:919–927. 27. Steenholdt C, Bendtzen K, Brynskov J, et al. Cut-off levels and diagnostic accuracy of infliximab trough levels and anti-infliximab antibodies in Crohn’s disease. Scand J Gastroenterol 2011;46:310–318. 28. Khanna R, Sattin BD, Afif W, et al. Review article: a clinician’s guide for therapeutic drug monitoring of infliximab in inflammatory bowel disease. Aliment Pharmacol Ther 2013;38:447–459. 29. Davidov Y, Ungar B, Bar-Yoseph H, et al. Association of induction infliximab levels with clinical response in perianal Crohn’s disease. J Crohns Colitis 2017;11:549–555. 30. Ungar B, Mazor Y, Weisshof R, et al. Induction infliximab levels among patients with acute severe ulcerative colitis compared with patients with moderately severe ulcerative colitis. Aliment Pharmacol Ther 2016; 43:1293–1299. 31. Yarur AJ, Jain A, Hauenstein SI, et al. Higher adalimumab levels are associated with histologic and endoscopic remission in patients with Crohn’s disease and ulcerative colitis. Inflamm Bowel Dis 2016;22:409–415. 32. Hanauer SB, Feagan BG, Lichtenstein GR, et al. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. Lancet 2002;359:1541–1549. 33. Colombel JF, Sandborn WJ, Rutgeerts P, et al. Adalimumab for maintenance of clinical response and remission in patients with Crohn’s disease: the CHARM trial. Gastroenterology 2007;132:52–65. 34. Levesque BG, Sandborn WJ, Ruel J, et al. Converging goals of treatment of inflammatory bowel disease from clinical trials and practice. Gastroenterology 2015; 148:37–51 e1. 35. Velayos FS, Kahn JG, Sandborn WJ, et al. A test-based strategy is more cost effective than empiric dose escalation for patients with Crohn’s disease who lose responsiveness to infliximab. Clin Gastroenterol Hepatol 2013;11:654–666. 36. Wolf D, D’Haens G, Sandborn WJ, et al. Escalation to weekly dosing recaptures response in adalimumabtreated patients with moderately to severely active ulcerative colitis. Aliment Pharmacol Ther 2014;40: 486–497. 37. Gibson DJ, Heetun ZS, Redmond CE, et al. An accelerated infliximab induction regimen reduces the need for early colectomy in patients with acute severe ulcerative colitis. Clin Gastroenterol Hepatol 2015;13:330–335. 38. Vande Casteele N, Ferrante M, Van Assche G, et al. Trough concentrations of infliximab guide dosing for patients with inflammatory bowel disease. Gastroenterology 2015;148:1320–1329. 39. Vaughn BP, Martinez-Vazquez M, Patwardhan VR, et al. Proactive therapeutic concentration monitoring of infliximab may improve outcomes for patients with inflammatory bowel disease: results from a pilot observational study. Inflamm Bowel Dis 2014;20:1996–2003.

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

AGA SECTION

-

2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520

22

AGA SECTION

2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580

Casteele et al

Gastroenterology Vol.

40. Adedokun OJ, Sandborn WJ, Feagan BG, et al. Association between serum concentration of infliximab and efficacy in adult patients with ulcerative colitis. Gastroenterology 2014;147:1296–1307. 41. Reinisch W, Colombel JF, Sandborn WJ, et al. Factors associated with short- and long-term outcomes oftherapy for crohn’s disease. Clin Gastroenterol Hepatol 2015;13:539–547. 42. Sharma S, Eckert D, Hyams JS, et al. Pharmacokinetics and exposure-efficacy relationship of adalimumab in pediatric patients with moderate to severe Crohn’s disease: results from a randomized, multicenter, phase-3 study. Inflamm Bowel Dis 2015;21:783–792. 43. Cornillie F, Hanauer SB, Diamond RH, et al. Postinduction serum infliximab trough level and decrease of C-reactive protein level are associated with durable sustained response to infliximab: a retrospective analysis of the ACCENT I trial. Gut 2014;63: 1721–1727. 44. Williet N, Boschetti G, Fovet M, et al. Association between low trough levels of vedolizumab during induction therapy for inflammatory bowel diseases with need for additional doses within 6 months. Clin Gastroenterol Hepatol 2016 Nov 24. pii: S1542–3565(16)31121-1, http://dx.doi.org/ 10.1016/j.cgh.2016.11.023 [Epub ahead of print]. 45. Adedokun OJ, Xu Z, Marano CW, et al. Pharmacokinetics and exposure-response relationship of golimumab in patients with moderately-to-severely active ulcerative colitis: results from Phase 2/3 PURSUIT Induction and Maintenance Studies. J Crohns Colitis 2017; 11:35–46. 46. Warman A, Straathof JWA, Derijks LJJ. Therapeutic drug monitoring of infliximab in inflammatory bowel disease patients in a teaching hospital setting: results of a prospective cohort study. Eur J Gastroenterol Hepatol 2015; 27:242–248. 47. Arias MT, Vande Casteele N, Vermeire S, et al. A panel to predict long-term outcome of infliximab therapy for patients with ulcerative colitis. Clin Gastroenterol Hepatol 2015;13:531–538. 48. Bortlik M, Duricova D, Malickova K, et al. Infliximab trough levels may predict sustained response to infliximab in patients with Crohn’s disease. J Crohns Colitis 2013;7:736–743. 49. Singh N, Rosenthal CJ, Melmed GY, et al. Early infliximab trough levels are associated with persistent remission in pediatric patients with inflammatory bowel disease. Inflamm Bowel Dis 2014;20:1708–1713. 50. Mazor Y, Almog R, Kopylov U, et al. Adalimumab drug and antibody levels as predictors of clinical and laboratory response in patients with Crohn’s disease. Aliment Pharmacol Ther 2014;40:620–628. 51. Ward MG, Kariyawasam VC, Mogan SB, et al. Clinical utility of measuring adalimumab trough levels and antibodies to adalimumab in patients with inflammatory bowel diseases. J Gastroenterol Hepatol 2013; 28:100–101. 52. Steenholdt C, Frederiksen MT, Bendtzen K, et al. Time course and clinical implications of development of binding and neutralizing antibodies against adalimumab

53.

54.

55.

56.

57.

58.

59.

60.

61.

62. 63.

64.

65.

66.

67.

-,

No.

-

in patients with inflammatory bowel disease. J Crohns Colitis 2015;9:S360–S361. Frederiksen MT, Ainsworth MA, Brynskov J, et al. Antibodies against infliximab are associated with increased risk of anti-adalimumab antibody development in patients with inflammatory bowel disease. Gastroenterology 2014;146:S238. Chaparro M, Guerra I, Iborra M, et al. Correlation between adalimumab serum levels and remission after the induction phase in crohn’s disease patients. Gastroenterology 2015;148:S107–S108. Vande Casteele N, FB, Vermeire S, Y, et al. Covariates influencing the exposure-response relationship for certolizumab pegol during the induction and maintenances phases in patients with Crohn’s disease. Am J Gastroenterol 2016;111:S262–S263. Detrez I, Dreesen E, Van Stappen T, et al. Variability in golimumab exposure: a ’real-life’ observational study in active ulcerative colitis. J Crohns Colitis 2016;10: 575–581. Vande Casteele N, Gils A, Singh S, et al. Antibody response to infliximab and its impact on pharmacokinetics can be transient. Am J Gastroenterol 2013; 108:962–971. Ungar B, Anafy A, Yanai H, et al. Significance of low level infliximab in the absence of anti-infliximab antibodies. World J Gastroenterol 2015;21:1907–1914. Ungar B, Chowers Y, Yavzori M, et al. The temporal evolution of antidrug antibodies in patients with inflammatory bowel disease treated with infliximab. Gut 2014; 63:1258–1264. Baert F, Noman M, Vermeire S, et al. Influence of immunogenicity on the long-term efficacy of infliximab in Crohn’s disease. N Engl J Med 2003;348:601–608. Dewit O, Moreels T, Baert F, et al. Limitations of extensive TPMT genotyping in the management of azathioprine-induced myelosuppression in IBD patients. Clin Biochem 2011;44:1062–1066. Lennard L. Implementation of TPMT testing. Br J Clin Pharmacol 2014;77:704–714. Booth RA, Ansari MT, Loit E, et al. Assessment of thiopurine S-methyltransferase activity in patients prescribed thiopurines: a systematic review. Ann Intern Med 2011;154:814–823, W-295-8. Colombel J, Ferrari N, Debuysere H, et al. Genotypic analysis of thiopurine S-methyltransferase in patients with Crohn’s disease and severe myelosuppression during azathioprine therapy. Gastroenterology 2000; 118:1025–1030. Lewis JD, Abramson O, Pascua M, et al. Timing of myelosuppression during thiopurine therapy for inflammatory bowel disease: implications for monitoring recommendations. Clin Gastroenterol Hepatol 2009;7: 1195–1201. Food and Drug Administration. Pediatric Oncology Subcommittee of the Oncologic Drugs Advisory Committee. Silver Spring, MD: Food and Drug Administration, 2003. Osterman MT, Kundu R, Lichtenstein GR, et al. Association of 6-thioguanine nucleotide levels and inflammatory

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

Q11

2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640

2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700

2017

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

AGA Review of Therapeutic Drug Monitoring in IBD

bowel disease activity: a meta-analysis. Gastroenterology 2006;130:1047–1053. Moreau AC, Paul S, Del Tedesco E, et al. Association between 6-thioguanine nucleotides levels and clinical remission in inflammatory disease: a meta-analysis. Inflamm Bowel Dis 2014;20:464–471. Coenen MJH, De Jong DJ, Van Marrewijk CJ, et al. Identification of patients with variants in TPMT and dose reduction reduces hematologic events during thiopurine treatment of inflammatory bowel disease. Gastroenterology 2015;149:907–917. Newman WG, Payne K, Tricker K, et al. A pragmatic randomized controlled trial of thiopurine methyltransferase genotyping prior to azathioprine treatment: the TARGET study. Pharmacogenomics 2011;12: 815–826. Sayani FA, Prosser C, Bailey RJ, et al. Thiopurine methyltransferase enzyme activity determination before treatment of inflammatory bowel disease with azathioprine: effect on cost and adverse events. Can J Gastroenterol 2005;19:147–151. Roy LM, Zur RM, Uleryk E, et al. Thiopurine Smethyltransferase testing for averting drug toxicity in patients receiving thiopurines: a systematic review. Pharmacogenomics 2016;17:633–656. Takatsu N, Matsui T, Murakami Y, et al. Adverse reactions to azathioprine cannot be predicted by thiopurine S-methyltransferase genotype in Japanese patients with inflammatory bowel disease. J Gastroenterol Hepatol 2009;24:1258–1264. Mottet C, Schoepfer AM, Juillerat P, et al. Experts opinion on the practical use of azathioprine and 6-mercaptopurine in inflammatory bowel disease. Inflamm Bowel Dis 2016;22:2733–2747. Thompson AJ, Newman WG, Elliott RA, et al. The costeffectiveness of a pharmacogenetic test: a trial-based evaluation of TPMT genotyping for azathioprine. Value Health 2014;17:22–33. Haines ML, Ajlouni Y, Irving PM, et al. Clinical usefulness of therapeutic drug monitoring of thiopurines in patients with inadequately controlled inflammatory bowel disease. Inflamm Bowel Dis 2011;17:1301–1307. Yarur AJ, Kubiliun MJ, Czul F, et al. Concentrations of 6-thioguanine nucleotide correlate with trough levels of infliximab in patients with inflammatory bowel disease on combination therapy. Clin Gastroenterol Hepatol 2015; 13:1118–1124. Cuffari C, Hunt S, Bayless T. Utilisation of erythrocyte 6-thioguanine metabolite levels to optimise azathioprine therapy in patients with inflammatory bowel disease. Gut 2001;48:642–646. Goldberg R, Moore G, Cunningham G, et al. Thiopurine metabolite testing in inflammatory bowel disease. J Gastroenterol Hepatol 2016;31:553–560. Kennedy NA, Asser TL, Mountifield RE, et al. Thiopurine metabolite measurement leads to changes in management of inflammatory bowel disease. Intern Med J 2013; 43:278–286.

23

81. Mardini HE, Arnold GL. Utility of measuring 6methylmercaptopurine and 6-thioguanine nucleotide levels in managing inflammatory bowel disease patients treated with 6-mercaptopurine in a clinical practice setting. J Clin Gastroenterol 2003;36: 390–395. 82. Smith M, Blaker P, Patel C, et al. The impact of introducing thioguanine nucleotide monitoring into an inflammatory bowel disease clinic. Int J Clin Pract 2013; 67:161–169. 83. Dassopoulos T, Dubinsky MC, Bentsen JL, et al. Randomised clinical trial: individualised vs weight-based dosing of azathioprine in Crohn’s disease. Aliment Pharmacol Ther 2014;39:163–175. 84. Reinshagen M, Schütz E, Armstrong VW, et al. 6-thioguanine nucleotide-adapted azathioprine therapy does not lead to higher remission rates than standard therapy in chronic active Crohn disease: results from a randomized, controlled, open trial. Clin Chem 2007;53: 1306–1314. 85. Van Asseldonk DP, Sanderson J, de Boer NKH, et al. Difficulties and possibilities with thiopurine therapy in inflammatory bowel disease—Proceedings of the first Thiopurine Task Force meeting. Dig Liver Dis 2011; 43:270–276. 86. Pavlidis P, Stamoulos P, Abdulrehman A, et al. Longterm safety and efficacy of low-dose azathioprine and allopurinol cotherapy in inflammatory bowel disease: a large observational study. Inflamm Bowel Dis 2016; 22:1639–1646. 87. Hoentjen F, Seinen ML, Hanauer SB, et al. Safety and effectiveness of long-term allopurinol-thiopurine maintenance treatment in inflammatory bowel disease. Inflamm Bowel Dis 2013;19:363–369.

Reprint requests Address requests for reprints to: Chair, Clinical Guidelines Committee, Q2 American Gastroenterological Association, National Office, 4930 Del Ray Avenue, Bethesda, Maryland 20814. e-mail: [email protected]; fax: ---. Q3 Acknowledgments The authors sincerely thank Kellee Kaulback, Medical Information Officer, Health Quality Ontario, for helping in the literature search for this technical review. The authors sincerely thank the following investigators for sharing unpublished data from their studies: Omoniyi Adedokun (Janssen Research and Development, Spring House, Pennsylvania), Martin Bortlik (Charles University, Prague, Czech Republic), Marc Ferrante (KU Leuven, Leuven, Belgium), Peter Irving (St Thomas Hospital, London, United Kingdom), Viraj Kariyawasam (University of Western Sydney, Australia), Konstantinos Papamichael (Beth Israel Deaconess Medical Center, Boston, Massachusetts and KU Leuven, Leuven, Belgium), Casper Steenholdt (Herlev University Hospital, Herlev, Denmark), Niels Vande Casteele (University of California San Diego, La Jolla, California), Mark Ward (Alfred Hospital, Melbourne, Australia), and Andres Yarur (Medical College of Wisconsin, Milwaukee, Wisconsin). Niels Vande Casteele and Hans Herfarth contributed equally to this work. Q10 Conflicts of interest The authors disclose no conflicts.

Q4

Funding Dr Singh is supported by the National Library of Medicine training grantQ5 T15LM011271, American College of Gastroenterology and Crohn’s and Colitis Foundation. Dr Herfarth is supported by National Institutes of Health grant 5U01DK092239, Crohn’s and Colitis Foundation, and the Broad Medical Research Program.

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

AGA SECTION

-

2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760

23.e1

2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820

Casteele et al

Gastroenterology Vol.

No.

Search Strategy

13

To inform evidence pertaining to these focused questions, a systematic literature search of multiple electronic databases on TDM in IBDs was conducted by an experienced medical librarian using a combination of controlled vocabulary supplemented with keywords, with input from the technical review authors. The search was conducted from inception to March 6, 2016, and the databases included: Ovid Medline In-Process & Other Non-Indexed Citations, Ovid MEDLINE, Ovid EMBASE and Ovid Cochrane Database of Systematic Reviews (detailed search strategy listed below). Technical review context experts and methodologist (SS) independently reviewed the title and abstract of studies identified in the search to exclude studies that did not address the focused question based on prespecified inclusion and exclusion criteria. The full text of the remaining articles was examined to determine whether it contained relevant information, and relevant articles at end of this process were selected for each question. Conflicts in study selection at this stage were resolved by consensus, referring back to the original article in consultation with clinical content experts. This search was supplemented with a recursive search of the bibliographies of recently published systematic reviews on this topic, to identify any additional studies. We also reviewed conference proceedings from major gastroenterology conferences from 2011 through 2016, and contacted experts in the field for any potential unpublished studies. We restricted our search to English language and human studies. Filters were applied to exclude editorials, letters to the editor and case reports.

Electronic Search Strategy

1 2 3 4 5 6 7 8

9 10 11 12

exp Inflammatory Bowel Diseases/ use mesz exp inflammatory bowel disease/ use emez (ibd or inflammatory bowel disease* or colitis or crohn*).ti,ab. (((enteritis or leitis or ileitides) adj2 (terminal or regional or granulomatous)) or ileocolitis).ti,ab. or/1-4 exp Antibodies, Monoclonal/ use mesz exp monoclonal antibody/ use emez (Infliximab or remicade or adalimumab or humira or certolizumab pegol or Cimzia or golimumab or Simponi or vedolizumab or Entyvio).ti,ab. or/6-8 exp Azathioprine/ exp 6-Mercaptopurine/ use mesz exp mercaptopurine/ use emez

14 15 16 17 18 19 20 21 22 23 24 25 26

27 28

29

30

Search date: March 6, 2016 Databases searched: Ovid MEDLINE 1946 to February Week 4 2016, Ovid MEDLINE In-Process & Other Non-Indexed Citations March 04, 2016, Embase 1980 to 2016 Week 10, Wiley Cochrane.

Searches

No.

-

Continued

Supplementary Methods

No.

-,

Results 64,883 100,911 216,329

31 32 33 34 35 36 37 38 39

Searches (6-Mercaptopurine or 6-MP or Purinethol or Purixan or azothioprine or Azathioprine or Azasan or Imuran).ti,ab. (thiopurine adj2 (therap* or treatment*)).ti,ab. or/10-14 5 and (9 or 15) exp Drug Monitoring/ exp Guanine Nucleotides/ use mesz exp guanine nucleotide/ use emez exp Thioguanine/ use mesz exp tioguanine/ use emez exp Methyltransferases/ use mesz exp thiopurine methyltransferase/ use emez exp Thionucleotides/ use mesz exp nucleotide/ use emez ((anti?drug* or anti?infliximab or anti?adalimumab or anti?Certolizumab pegol or anti?vedolizumab or anti?golimumab or Human Anti?chimeric or HACA or anti?remicade or anti?humira or anti?cimzia or anti?simponi or anti?entyvio) adj2 anti?bod*).ti,ab. (therapeutic adj2 monitor*).ti,ab. (thiopurine methyltransferase or tpmt or thiopurine S? methyltransferase or 6?thioguanine* or 6?tgn or 6?methylmercaptopurine ribonucleotide* or 6? MMP or 6?MMPR).ti,ab. ((drug* or medication* or biologic* or infliximab or adalimumab or Certolizumab pegol or vedolizumab or golimumab or thiopurine* or 6? Mercaptopurine or Azathioprine or anti?bod* or metabolite* or metabolism or anti?bod*) adj2 (concentration* or measure* or level* or monitor* or test*)).ti,ab. (trough or assay* or immuno?assay* or radioimmuno? assay* or elisa or phenotype or gentotype).ti,ab. or/17-30 16 and 31 Case Reports/ or Comment.pt. or Editorial.pt. or Letter.pt. Case Report/ or Comment/ or Editorial/ or Letter/ animals/ not (humans/ and animals/) 32 not (33 or 34 or 35) limit 36 to english language limit 37 to yr ¼ “2000 -Current” remove duplicates from 38

Results 38,217

919 117,795 34,290 59,744 50,692 4993 2449 7780 35,464 2023 10,136 486,603 657

19,666 2539

325,351

2,696,842 3,504,924 5282 4,456,599 6,290,700 5,310,359 4795 4605 4412 3659

Wiley Cochrane

2753 246,619 192,104 396,863 33,037

592,430 89,349 18,207 21,019

ID Search #1 MeSH descriptor: [Inflammatory Bowel Diseases] explode all trees #2 (ibd or inflammatory bowel disease* or colitis or crohn*):ti,ab #3 (((enteritis or leitis or ileitides) near/2 (terminal or regional or granulomatous)) or ileocolitis):ti,ab #4 #1 or #2 or #3 #5 MeSH descriptor: [Antibodies, Monoclonal] explode all trees

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

Hits 2246 4129 25 4261 6272

2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880

-

2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940

2017

AGA Review of Therapeutic Drug Monitoring in IBD

Continued #6 (Infliximab or remicade or adalimumab or humira or certolizumab pegol or Cimzia or golimumab or Simponi or vedolizumab or Entyvio):ti,ab #7 #5 or #6 #8 MeSH descriptor: [Azathioprine] explode all trees #9 MeSH descriptor: [6-Mercaptopurine] explode all trees #10 (6-Mercaptopurine or 6-MP or Purinethol or Purixan or azothioprine or Azathioprine or Azasan or Imuran):ti,ab #11 (thiopurine near/2 (therap* or treatment*)):ti,ab #12 #8 or #9 or #10 or #11 #13 #4 and (#7 or #12) #14 MeSH descriptor: [Drug Monitoring] explode all trees #15 MeSH descriptor: [Guanine Nucleotides] explode all trees #16 MeSH descriptor: [Thioguanine] explode all trees #17 MeSH descriptor: [Methyltransferases] explode all trees #18 MeSH descriptor: [Thionucleotides] explode all trees #19 ((anti*drug* or anti*infliximab or anti*adalimumab or anti*Certolizumab pegol or anti*vedolizumab or anti*golimumab or Human Anti*chimeric or HACA or anti*remicade or anti*humira or anti*cimzia or anti*simponi or anti*entyvio) near/2 anti*bod*):ti,ab #20 (therapeutic near/2 monitor*):ti,ab #21 (thiopurine methyltransferase or tpmt or thiopurine S-methyltransferase or 6*thioguanine* or 6*tgn or 6*methylmercaptopurine ribonucleotide* or 6*MMP or 6*MMPR):ti,ab #22 ((drug* or medication* or biologic* or infliximab or adalimumab or Certolizumab pegol or vedolizumab or golimumab or thiopurine* or 6*Mercaptopurine or Azathioprine or anti*bod* or metabolite* or metabolism or anti*bod*) near/2 (concentration* or measure* or level* or monitor* or test*)):ti,ab #23 (trough or assay* or immuno*assay* or radioimmuno*assay* or elisa or phenotype or gentotype).ti,ab. #24 #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 #25 #13 and #24 Publication Year from 2000 to 2016

2020

7369 1099 1310 1730 22 2238 805 1319 272 175 284 59 30

371 58

13205

293

15351 91

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

23.e2

2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000

23.e3

3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060

Casteele et al

Gastroenterology Vol.

-,

No.

-

Supplementary Table 1.Main Characteristics of Assays Currently Available in the United States Assay ELISA

HMSA

LC-MS/MS ECLIA

Target

Phase

LLOQ

Drug tolerant

Provider

Adalimumab Anti-adalimumab Abs Infliximab Anti-infliximab Abs Certolizumab pegol Anti-certolizumab pegol Abs Vedolizumab Anti-vedolizumab Abs Ustekinumab Anti-ustekinumab Abs Anti-infliximab Abs Adalimumab Anti-adalimumab Abs Infliximab Anti-infliximab Abs Vedolizumab Anti-vedolizumab Abs Infliximab Adalimumab Anti-adalimumab Abs Infliximab Anti-infliximab Abs

Solid Solid Solid Solid Solid Solid Solid Solid Solid Solid Solid Fluid Fluid Fluid Fluid Fluid Fluid Fluid Fluid Fluid Fluid Fluid

0.1 mg/mL 10 ng/mL 0.1 mg/mL 10 ng/mL 0.4 mg/mL 10 ng/mL 2 mg/mL 35 ng/mL 0.2 mg/mL 5 ng/mL 50 U/mL 1.6 mg/mL 1.7 U/mL 0.98 mg/mL 3.13 U/mL 1.6 mg/mL 1.6 U/mL 1.0 mg/mL 0.6 mg/mL 25 ng/mL 0.4 mg/mL 22 ng/mL

/ No / No / No / No / No No / Yes / Yes / Yes / / Yes / Yes

Miraca Miraca Miraca Miraca Miraca Miraca Miraca Miraca Miraca Miraca Mayo Prometheus Prometheus Prometheus Prometheus Prometheus Prometheus Mayo Esoterix Esoterix Esoterix Esoterix

ADAb, anti-drug antibody; LC-MS/MS, liquid chromatography tandem mass spectrometry; LLOQ, lower limit of quantification.

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120

3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 Paul, 201323 (prospective)

Drug

Inclusion

Intervention

52 (34 CD and 18 UC) Infliximab CD: CDAI 220 AND CRP >10 mg/L OR 10 mg/kg q8 wk 5 mg/kg q8 wk FC >450 mg/g UC/IC: SCCAI 3 AND endoscopic Mayo score >1 82 (46 CD and 36 UC) Adalimumab CD: CDAI 220 AND FC >450 mg/gUC: Mayo clinic 40 mg EW 40 mg EOW score >5 AND endoscopic Mayo score >1 247 patients (199 CD, Maintenance Clinical worsening by the treating physician, or definite Dose intensification/ 48 UC or IBDinfliximab inflammatory loss of response if clinical worsening switch anti-TNF Maintenance agent/expectant unclassified) was associated with either increased CRP or FC, adalimumab active endoscopic inflammation, fistula discharge, management or compatible imaging studies

Therapeutic trough

Low ADAbs

High ADAbs

2 mg/mL

>10 ng/mL and >200 ng/mL <200 ng/mL

4.9 mg/mL

>10 ng/mL

3.8 mg/mL 4.5 mg/mL

>4 mg/mL equivalents >9 mg/mL equivalents

CDAI, Crohn’s disease activity index; EOW, every other week; EW, every week; FC, fecal calprotectin; IC, indeterminate colitis; SCCAI, Simple Clinical Colitis Activity Index.

AGA Review of Therapeutic Drug Monitoring in IBD

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

Roblin, 201425 (prospective) Yanai, 201524 (retrospective)

Patients, n

2017

First author, year

-

Supplementary Table 2.Characteristics of Cohort Studies Evaluating the Impact of Reactive Therapeutic Drug Monitoring for Anti-Tumor Necrosis Factor
a Agents

23.e4

3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240

23.e5

3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300

Casteele et al

Gastroenterology Vol.

-,

No.

-

Supplementary Table 3.Proportion of Patients With Ulcerative Colitis or Crohn’s Disease, Not in Remission, at Different Infliximab Cutoffs in Cross-Sectional or Cohort Studies Which Reported Outcomes at Several Predefined Cutoffs Patients not in clinical remission UC Infliximab trough concentration threshold, mg/mL 1 3 5 7 10

% (95% CI) 32.7 16.6 10.1 3.6 2.8

(16.8
54.0) (7.1
34.4) (4.3
21.9) (0.6
19.2) (0.2
35.3)

CD No. of studies 3 3 3 3 2

% (95% CI) 19.7 11.8 6.7 4.5 3.5

(13.9
27.1) (7.7
17.7) (3.7
11.7) (1.1
16.0) (1.9
6.4)

No. of studies 3 3 3 2 2

Supplementary Table 4.Metabolite-Directed Algorithm for Clinical Strategies in Patients With Inflammatory Bowel Disease and an Inadequate Response to Thiopurine Therapy (as Proposed in Haines et al76) 6-TGN level

6-MMP level

Therapeutic (230
450 pmol/8  108 RBCs) Low (<230 pmol/8  108 RBCs)

High (>450 pmol/8  108 RBCs) Low (<230 pmol/8  108 RBCs)

Normal or high Low or normal

High or normal High (>5700), with 6-MMP/6TGN ratio >11

Interpretation Thiopurine-refractory, appropriately dosed Underdosed, or noncompliant (if both 6-TGN and 6-MMP low, and questioning patient) Thiopurine-refractory, overdosed 6-MMP shunter

Proposed management strategy Change therapy; if thiopurine continued, no change in dose Increase dose; if noncompliant, then education about compliance

Change therapy Change therapy or add allopurinol and reduce thiopurine dose to 25%

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360

-

3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420

2017

AGA Review of Therapeutic Drug Monitoring in IBD

23.e6

Supplementary Table 5.Impact of Thiopurine Metabolite Testing on Effectiveness and Safety in Observational Studies Reason for TDM First author, year

Patients in study, n

Inadequate response

Mardini, 200381

44

Kennedy, 201380

AEs

Outcomes based on TDM results

31

13 (7 abnormal LFTs, 2 leukopenia, 4 anemia)

98

80

18 (6 abnormal LFTs, 5 leukopenia, 7 other AEs)

Smith, 201382

189

53

NR

Goldberg, 201679 Total

151 545

116 343

30a 61

Treatment change reported only for subtherapeutic: 11 of 19 (58%) remission after dose increase Resolution of AE: overall, 6 of 13 (46%); LFTs: 5 of 7 (71%); leukopenia: 1 of 2 (50%) Improved clinical outcome (which includes increasing thiopurines, adding allopurinol, switch of therapy including surgery): 44 of 80 (55%) Resolution AE after TDM: overall, 5 of 18 (28%); LFTs: 1 of 6 (17%); leukopenia: 2 of 5 (40%); other AEs: 2 of 7 (29%). Improvement of 18 of 20 (90%) with treatment change based on 6-TGN vs 7 of 21 (33%) in treatment changes not based on 6-TGN levels Change of therapy in 78 of 116 (67%) and 18 of 30 (60%)a

AE, adverse event; LFT, liver function test; NR, not reported. a Includes patients with side effects and other indications for drug monitoring.

PGL 5.5.0 DTD
YGAST61322_proof
31 July 2017
5:29 pm
ce

3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480