Effect of Nucleoside and Nucleotide Analogues on Renal Function in Patients With Chronic Hepatitis B Virus Monoinfection

All studies published in Clinical Gastroenterology and Hepatology are embargoed until 3PM ET of the day they are published as corrected proofs on-line. Studies cannot be publicized as accepted manuscripts or uncorrected proofs.

Clinical Gastroenterology and Hepatology 2014;-:-–-

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

Effect of Nucleoside and Nucleotide Analogues on Renal Function in Patients With Chronic Hepatitis B Virus Monoinfection Q32

Vincent Mallet, Michaël Schwarzinger, Anaïs Vallet-Pichard, Hélène Fontaine, Marion Corouge, Philippe Sogni, and Stanislas Pol

Q2

Institut Cochin, Université Paris Descartes, Sorbonne Paris Cité, Unité Mixte de Recherche, Paris, France; Centre National de la Recherche Scientifique, Unités Mixtes de Recherche, Paris, France; Institut National de la Santé et de la Recherche Médicale Unité, Paris, France; Assistance Publique-Hôpitaux de Paris, Groupe Hospitalier Cochin Saint-Vincent de Paul, Département d’Hépatologie, Paris, France; Translational Health Economics Network, Paris, France

Q8

BACKGROUND & AIMS:

There is controversy regarding whether nucleos(t)ide analogues contribute to renal impairment in patients with chronic hepatitis B virus (HBV) infection. We analyzed changes in renal function in patients with chronic HBV infection and whether these were associated with treatment or comorbidities.

METHODS:

We performed a longitudinal observational study to investigate factors associated with renal function in 214 patients (median age, 43 y; 69.2% men) with compensated chronic HBV monoinfection treated with 343 lines of nucleos(t)ide analogues (210 monotherapies, 133 combinations) between 1990 and 2012 (median time, 2.4 y) in France. A linear mixed-effect model was used to model variations of estimated glomerular filtration rate (eGFR, computed with the Chronic Kidney Disease Epidemiology Collaboration formula), adjusting for age, sex, geographic origin, initial liver fibrosis, level of HBV DNA, and an eGFR less than 90 mL/min/ 1.73 m2.

RESULTS:

The eGFR decreased in patients given adefovir dipivoxil as monotherapy or in a combination (P < .0001 and P < .002, respectively), and remained stable in patients given lamivudine, tenofovir disoproxil fumarate, or entecavir. The eGFR decreased in patients with a baseline eGFR of less than 90 mL/min/1.73 m2, regardless of treatment. The eGFR remained stable or increased, regardless of treatment, in patients with a baseline eGFR of 90 mL/min/1.73 m2 or greater and with an initial HBV DNA level of 100,000 IU/mL or greater. Patients born in areas of high endemicity of HBV were more prone to increases in eGFR with treatment.

CONCLUSIONS:

In a real-life study, the eGFR remained stable or increased over time in patients with chronic HBV monoinfection with a baseline eGFR of 90 mL/min/1.73 m2 or higher and treated with tenofovir disoproxil fumarate or entecavir. Patients born in an area of high endemicity of HBV who had initial levels of HBV DNA of 100,000 IU/mL or greater were more likely to have an increased eGFR with treatment.

Keywords: BeSafe Study; Chronic Kidney Disease; Antiviral; Drug; Mixed Linear Model. Q9 Q10 Q11

here is a relationship between chronic hepatitis B virus (HBV) infection and chronic kidney disease (CKD). Chronic HBV infection is a cause of CKD in areas of the world with a high HBV prevalence, mainly through deposition of immune complexes in the kidney.1 HBVinduced CKD usually improves with spontaneous or treatment-induced control of viral replication.2,3 CKD also frequently is reported in patients with chronic HBV infection from industrialized countries. The mechanism of these renal abnormalities has been poorly characterized, mainly because of potential cofounding factors, including old age, diabetes mellitus and the

T

metabolic syndrome, high blood pressure, human immunodeficiency virus (HIV) or hepatitis C virus co-infections, nephrotoxic drugs, and end-stage liver disease. Abbreviations used in this paper: ADV, adefovir dipivoxil; CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; ETV, entecavir; HBV, hepatitis B virus; HIV, human immunodeficiency virus; IQR, interquartile range; LAM, lamivudine; LdT, telbivudine; NUC, nucleotide or nucleoside analogue; TDF, tenofovir disoproxil fumarate. © 2014 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2014.11.021

FLA 5.2.0 DTD
YJCGH54078_proof
5 December 2014
3:27 pm
ce DVC

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

2

117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174

Q12

Q13

Mallet et al

Clinical Gastroenterology and Hepatology Vol.

Treatment of HBV infection includes nucleoside (lamivudine [LAM], telbivudine [LdT], or entecavir [ETV]) or nucleotide (adefovir dipivoxil [ADV] or tenofovir disoproxil fumarate [TDF]) analogues (NUCs) for an indefinite duration. Second-generation NUCs are recommended in industrialized countries.4 Renal excretion is the primary route of elimination of NUCs.5 All NUCs harbor a dose-dependent kidney toxicity by various mechanisms, including alterations in renal tubular transporters, apoptosis, and mitochondrial toxicity.6 Registrations trials on NUCs showed a safe renal profile in patients without risk factors of CKD, with some finding an increase in the estimated glomerular filtration rate (eGFR) over time.7 In real-life studies, results regarding renal tolerance of NUCs are controversial, with some studies finding either a decrease or an increase in eGFR over time under various classes of NUCs.4,8–11 None of these studies took into account other potential cofounding risk factors of CKD. The aim of this retrospective and prospective longitudinal, observational, Besafe study was to model renal function under NUCs in chronic HBV infection, taking into account known risk factors of CKD, including NUC exposure.

Methods Study Design

Q14

Q15

Q16

We selected from an integrated database of 2485 patients with HBV a retrospective cohort of patients with chronic HBV infection who received specific treatment between 1990 and 2012 at a single hepatology unit in a tertiary care center in France (Appendix Figure 1). We included patients who met the following criteria: chronic HBV infection treated with NUC; absence of hepatitis C and hepatitis D co-infection; absence of liverrelated complication, including ascites, hepatic encephalopathy, variceal bleeding, spontaneous bacterial peritonitis, hepatocellular carcinoma, or liver transplantation; absence of immunosuppression, including HIV-associated infection, long-term hemodialysis, organ transplantation, or immunosuppressive therapy; absence of solid cancer or leukemia; and a Charlson/Deyo comorbidity index of 1 or less.12 All patients in this cohort underwent a prospective follow-up evaluation by a senior hepatologist at least every 6 months with a clinical and biological evaluation at each visit that included measurement of blood aminotransferase level, bilirubin level, creatinine level, and HBV viral load for all patients, and albumin level, prothrombin time, platelet level, and blood phosphorus level for most of the patients. Nucleoside and nucleotide analogue dosing was adapted to creatinine clearance based on approved special product characteristics and according to current guidelines.4 The eGFR rate was computed with the Chronic Kidney Disease Epidemiology Collaboration equation and chronic

-,

No.

-

kidney disease was defined according to practice guidelines.13 Liver fibrosis was defined according to invasive (histologic) and/or clinical (gastroscopy, ultrasound, liver stiffness) findings.

Outcome Measures The BeSafe study was designed to model the evolution of renal function over time in real-life patients with chronic HBV infection treated with NUCs. The primary end point was to evaluate the effect of NUCs (or combination of nucleoside and nucleotide analogues) on eGFR over time. The effect of LAM, ETV, ADV, TDF, and of nucleoside/nucleotide combination on eGFR was evaluated as a secondary end point.

Data Collection All patients’ records were dematerialized in an integrated data warehouse and the contents were indexed (biological data, radiologic reports, written file, letters). A senior physician coded all diagnoses and comorbidities according to the International Classification of Diseases, 10th revision. All treatments and dates of initiation, stops, add-ons, and switches were recorded. Biological information was collected prospectively by the means of direct links with the hospital’s centralized biological database and optical character recognition of any dematerialized externalized biological information. Quality control of every optical character recognition was performed by viewing the original report and correcting the information automatically imported in the database.

Statistical Analysis Continuous values were summarized by median and interquartile ranges (IQRs) or means and 95% confidence intervals (CIs), and categoric variables by proportions. Time was calculated from the date of treatment initiation to the date of measure. The median time period between 2 measures was 12 weeks (IQR, 8–24 wk). The median number of measures of eGFR per patient was 7 (IQR, 3–14). The median time from the first to the last measure was 30 months (IQR, 8–91 mo). The differences between groups were assessed with the Fisher exact test, the Mann–Whitney U test for unrelated samples, and the paired t test for related samples. The impact of treatments on eGFR was estimated with a linear mixed-effects model accounting for repeated measures with a diagonal covariance structure for the residuals for each line of treatment, with some patients being treated with several lines. Age, sex, and the presence of the following factors at baseline: geographic origin, extensive fibrosis or cirrhosis before treatment, diabetes, obesity or hyperlipemia, high blood pressure (defined as receiving a drug for blood pressure), high

FLA 5.2.0 DTD
YJCGH54078_proof
5 December 2014
3:27 pm
ce DVC

Q17

175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232

-

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

2014

Renal Effect of NUC in Chronic Hepatitis B

baseline viral load (HBV DNA
100,000 IU/mL), baseline eGFR level less than 90 mL/min/1.73 m2, and exposure to nucleosides and/or nucleotides were entered as fixed effects in the first model. In the second model, exposition to LAM, ADV, TDF, ETV, or to a combination of nucleos(t)ide analogues was entered as a fixed effect; and geographic origin, extensive fibrosis or cirrhosis before treatment, high viral load (HBV DNA
100,000 IU/mL), and eGFR less than 90 mL/min/1.73 m2 at baseline were entered as random effects. In the third model, the interaction between exposition to ADV, TDF, or ETV, baseline eGFR less than 90 mL/min/1.73 m2, baseline viral load of 100,000 IU/mL or greater, and geographic origin were entered as fixed effects. Age, sex, and baseline liver fibrosis were entered as random effects. Because of the lack of clear association between smoking and CKD, this variable was not entered in the models.14 The estimates correspond to the mean variation of eGFR for binary variables or per person-year of exposure to a drug. All P values are 2-sided, and the type I error was set at 5%. All statistical analyses were performed using SPSS software, version 20 (IBM, BoisColombes, France).

Results Sample The cohort comprised 214 patients with chronic HBV infection without co-infection, without immunodepression, and without more than 1 life-threatening condition described in the Charlson/Deyo index, including

3

Q1

end-stage liver disease and hepatocellular carcinoma (Appendix Figure 1). The median age was 43 years; twothirds were men and one-fourth was of Asian ethnic origin. Approximately 10% of patients had 1 or more risk factors of CKD, including diabetes, obesity, hyperlipemia, and high blood pressure. An eGFR less than 90 mL/min/ 1.73 m2 was present in approximately 40% of patients before treatment initiation. The median HBV DNA level was 4.4 logIU/mL (IQR, 2.9–6.6 logIU/mL), and approximately half of the patients had direct (biopsy-proven) or indirect signs of extensive liver fibrosis (N ¼ 63; 29.4%) or cirrhosis (N ¼ 35; 16.4%) before treatment (Table 1).

Unadjusted Measure of Estimated Glomerular Filtration Rate During Treatment All patients were treated for a median time period of 2.4 years (IQR, 0.6–6.0 y), with 343 different lines of NUC monotherapies or nucleoside/nucleotide combinations (Table 2). The median treatment time with a nucleoside, nucleotide, and nucleoside/nucleotide combination was 1.9 years (IQR, 0.8–3.1 y), 2.0 years (IQR, 1.0–3.0 y), and 3.3 years (IQR, 2.0–4.8 y), respectively (P < .0001 for multiple comparisons and P ¼ .478 when comparing nucleotide and nucleoside exposure). The median number of eGFR measures was 7 (IQR, 3–14) per patient. There were more measures with nucleotide treatment (6; IQR, 3–11) and nucleoside/nucleotide combination treatment (12; IQR, 5–22) than with nucleoside treatment (4; IQR, 2–7) (P < .0001). The median eGFR at baseline was 91 mL/min/1.73 m2 (IQR, 79–104 mL/min/1.73 m2), 92 mL/min/1.73 m2

Table 1. Characteristics of 214 HBV-Infected Patients Treated With Nucleoside and/or Nucleotide Analogue Characteristics (N ¼ 214) Median (IQR) age, y Male sex, n (%) Asian ethnic origin, n (%) African ethnic origin, n (%) Median (IQR) body mass index, kg/m2 Metabolic syndrome, n (%)a High blood pressure, n (%)b Extensive fibrosis or cirrhosis, n (%)c HBV DNA level (IQR), logIU/mL Median (IQR) AST level, IU/L Median (IQR) ALT level, IU/L Median (IQR) platelet count, 109 cells/L Median (IQR) prothrombin time, % of normal Median (IQR) blood phosphorus level, mmol/Le Median (IQR) eGFR,f mL/min/1.73 m2 Number of patients with an eGFR < 90 mL/min/1.73 m2, n (%)

At baseline 43 148 54 47 24.3 25 27 98 4.4 38 52 176 92 1 94.4 83

(33–54) (69.2) (25.2) (22.0) (21.0–27.3) (11.7) (12.6) (45.8) (2.9–6.6) (29–59) (32–107) (142–217) (86–100) (0.96–1) (83.7–110.2) (38.8)

At the last follow-up measure

P value

47 (37.8–56)

<.0001

<1.2 30 29 185 92 1 95.7 80

(1.2–1.2)d (24–37) (24–37) (150–232) (83–99) (0.92–1) (79.4–108.2) (37.4)

<.0001 <.001 <.001 <.01 .191 .651 .182 <.0001

ALT, alanine aminotransferase; AST, aspartate aminotransferase. a Includes obesity in 3 patients, diabetes in 8 patients (2 requiring insulin), and hyperlipemia in 18 patients. b High blood pressure was defined as receiving a drug for blood pressure. c The presence of extensive fibrosis or of cirrhosis was defined histologically according to the Metavir system or was defined clinically (liver stiffness, gastroscopy, ultrasound) when indirect signs of cirrhosis were present. d Thirty-five patients (11.3%) had definitive cirrhosis. The last HBV DNA measurement was missing in 53 (15.4%) patients. e The median blood phosphorus levels were 1.00 (IQR, 0.93–1.03) and 1.00 (IQR, 0.90–1.01) for patients treated with entecavir and tenofovir, respectively (P ¼ .403). f Computed with the Chronic Kidney Disease Epidemiology Collaboration equation.32

FLA 5.2.0 DTD
YJCGH54078_proof
5 December 2014
3:27 pm
ce DVC

Q29

291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348

4

349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406

Mallet et al

Clinical Gastroenterology and Hepatology Vol.

Table 2. The Different Lines of Treatment by Nucleoside and/ or Nucleotide Analogue Lines of treatment

Q30

Tenofovir disoproxil fumarate, n (%) Adefovir dipivoxil, n (%) Entecavir, n (%) Lamivudine, n (%) Association of nucleoside and nucleotide analogues, n (%)a Treatment periodsa Median (IQR) time treated with lamivudine, y Median (IQR) time treated with adefovir dipivoxil, y Median (IQR) time treated with tenofovir disoproxil fumarate*, y Median (IQR) time treated with entecavir, y Median (IQR) time treated with an association of nucleoside and nucleotide analogues, y Number of measures of eGFR per patientb Median (IQR) number of measures with lamivudine, n Median (IQR) number of measures with adefovir dipivoxil, n Median (IQR) number of measures with tenofovir disoproxil fumarate, n Median (IQR) number of measures with entecavir, n Median (IQR) number of measures with an association of nucleoside and nucleotide analogues, n

(20.4) (8.7) (17.8) (14.3) (38.8)

1.5 (0.7–4.3) 2.7 (0.9–5.0) 1.8 (1.0–2.7) 1.8 (0.5–3.0) 3.3 (2.0–4.8)

3 (2–5) 5 (2–9) 6 (3–11) 4 (2–8) 12 (5–22)

a Includes 66 (49.6%) associations of ADV/LAM, 32 (24.1%) associations of TDF/LAM, 15 (11.3%) associations of TDF/ETV, 14 (10.5%) associations of TDF/FTC, and 6 (4.5%) associations of ADV/ETV. b P < .0001 for multiple comparisons.

(IQR, 80–111 mL/min/1.73 m2), and 99 mL/min/1.73 m2 (IQR, 82–112 mL/min/1.73 m2) with nucleoside, nucleotide, and a combination of nucleoside/nucleotide analogue treatment, respectively (P ¼ .103). Other baseline characteristics were similar between the 3 groups. The median difference between the last follow-up evaluation measure and the baseline measure was 0 mL/min/ 1.73 m2 (IQR, -3 to 5 mL/min/1.73 m2), 0 mL/min/1.73 m2 (IQR, -5 to 4 mL/min/1.73 m2), and 1 mL/min/1.73 m2 (IQR, -6 to 10 mL/min/1.73 m2) for patients treated by a nucleoside (P ¼ .909), nucleotide (P ¼ .645), and a combination of a nucleoside/nucleotide analogue (P ¼ .423), respectively. The number of patients with an eGFR less than 90 mL/min/m2 decreased between the first and the last measure: 21 (25.3%) patients with CKD stage 2 on the first measure were CKD stage 1 on the last measure and 18 (13.7%) patients with CKD stage 1 were CKD stage 2 on the last measure. The blood phosphorus level was similar between the first and last measures, regardless of treatment type. Almost all patients had an undetectable serum HBV DNA level, a normal transaminase level, and an increased platelet count on the last follow-up measure (Table 1). Among cirrhotic patients, 17 (48.6%) had an increased platelet level of more than 50,000/mm3, suggesting histologic improvement.

No.

-

Table 3. Estimated Impact of Host and Viral Factors on eGFR Over the Study Period

N ¼ 343 70 30 61 49 133

-,

Factor Intercept Age, mL/min/1.73 m2 per year Male sex, mL/min/1.73 m2 Born in an area of high endemicity, mL/min/1.73 m2 Presence of diabetes, obesity, or hyperlipemia, mL/min/1.73 m2 Presence of high blood pressure, mL/min/1.73 m2 Presence of extensive liver fibrosis or cirrhosis before treatment, mL/min/1.73 m2 Baseline eGFR < 90 mL/min/ 1.73 m2, mL/min/1.73 m2 Baseline HBV DNA level
100,000 IU/mL, mL/min/1.73 m2 Exposure to nucleotide analogues, mL/min/1.73 m2 per year Exposure to nucleoside analogues, mL/min/1.73 m2 per year Exposure to a combination of nucleoside and nucleotide analogues, mL/min/1.73 m2 per year

Mean eGFR variation, mL/min/ 1.73 m2 P value 127.2 (121.6–132.8) -0.5 (-0.6 to -0.5) -3.5 (-5.0 to -1.9) 9.6 (8.0–11.3)

<.0001 <.0001 <.0001 <.0001

0.0 (-2.8 to 2.8)

.999

-0.01 (-3.4 to 2.7)

.811

2.3 (-3.9 to -0.6)

.008

-8.5 (-10.3 to -6.6)

<.0001

7.9 (6.2–9.6)

<.0001

-0.9 (-1.7 to -0.1)

.034

-0.1 (-0.6 to 0.5)

.771

-0.6 (-1.0 to -0.2)

.007

Q31

Four (1.9%) patients with a baseline eGFR of 90 mL/ min/1.73 m2 or greater had an eGFR less than 60 mL/ min/1.73 m2 on the last follow-up measure, including 1 patient with high blood pressure treated for 2.3 years with ETV, 1 patient with high blood pressure and hyperlipemia treated for 4.9 years with ETV, 1 patient with baseline cirrhosis treated for 3.6 years with ADV/LAM and for 2.5 years with TDF, and 1 patient without any identified risk factor of CKD treated for 2.8 years with LAM.

Adjusted Estimation of Estimated Glomerular Filtration Rate Variations During Treatment Renal function is influenced by age, sex, and by other risk factors of CKD such as diabetes, obesity, hyperlipemia, or high blood pressure.15 Liver fibrosis may impair renal function. HBV is a cause of CKD, especially in people from endemic areas (vertical and horizontal transmission of HBV in early childhood). Whether NUC exposure is toxic to the kidney is controversial. We first entered all of these variables as fixed effects in a linear mixed model accounting for repeated measures. Age, male sex, baseline eGFR less than 90 mL/ min/1.73 m2, and nucleotide or nucleotide/nucleoside exposure decreased eGFR over time (Table 3). Interestingly, geographic origin (born in an area of high endemicity), a high baseline viral load (HBV DNA
100,000

FLA 5.2.0 DTD
YJCGH54078_proof
5 December 2014
3:27 pm
ce DVC

407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464

-

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

2014

Renal Effect of NUC in Chronic Hepatitis B

5

Table 4. Estimated Yearly Impact of Nucleoside and Nucleotide Analogue Monotherapies and Associations of Nucleoside and Nucleotide Analogues With eGFR During the Study Period

Estimates of fixed effectsa Intercept Mean eGFR variation with adefovir dipivoxil Mean eGFR variation with tenofovir disoproxil fumarate Mean eGFR variation with lamivudine Mean eGFR variation with entecavir Mean eGFR variation with nucleoside and nucleotide analogue association

Number of measures

Number of patients

Median exposure, y

Mean eGFR variation, mL/min/1.73 m2 per year (95% CI)

P value

167 370

19 47

2.9 2.5

121 (100–142) -2.2 (-3.3 to -1.1) 0.6 (-0.8 to 1.94)

<.0001 <.0001 .420

152 249 1053

36 45 48

2.9 2.5 3.2

-0.06 (-0.7 to 0.6) -0.1 (-1.5 to 1.3) -0.8 (-1.3 to -0.3)

.853 .872 .002

a Mixed model adjusted for age, sex, geographic origin, baseline liver fibrosis (extensive fibrosis or cirrhosis), baseline HBV DNA level less than 100,000 IU/mL, and baseline eGFR less than 90 mL/min/1.73 m2.

IU/mL), and the presence of extensive liver fibrosis or cirrhosis before treatment were protective. An increase in platelet count of 50,000/mm3 or greater was associated with an independent increase in eGFR when this variable was added in the model. High blood pressure and the metabolic syndrome decreased eGFR over time when age and baseline eGFR less than 90 mL/min/1.73 m2 were not entered in the model. We then built a model taking into account NUC exposure over time as fixed factors and entered as random effects the previously identified factors, including age, sex, geographic origin, baseline liver fibrosis (extensive fibrosis or cirrhosis), baseline HBV DNA level of 100,000 IU/mL or greater, and baseline eGFR (<90 mL/min/1.73 m2). We observed a decrease of eGFR with ADV treatment and with nucleotide/ nucleoside combinations. Among all combinations, only ADV/LAM was nephrotoxic (not shown). The eGFR remained stable with LAM, ETV, and TDF over the study period (Table 4). No interaction was detected with previous ADV exposure. We then studied the interaction between ADV, TDF, and ETV exposure, baseline viral load, and geographic origin in patients with a baseline eGFR of 90 mL/min/ 1.73 m2 or greater with all other factors entered in the model as random effects. In patients born in a nonendemic area with a baseline HBV DNA level of 100,000 IU/mL or greater, the mean eGFR variation was 1.5 (95% CI, 0.17–2.8), 4.7 (95% CI, 2.5–6.9), 3.3 (95% CI, 1.76–4.9) mL/min/1.73 m2 per year with ADV, ETV, and TDF treatment, respectively. In patients born in a nonendemic area with a baseline HBV DNA level of 100,000 IU/mL or greater, the eGFR remained stable with ADV and ETV treatment, and increased with TDF treatment (P ¼ .022).

Discussion The BeSafe observational study was designed to assess renal function variations of chronic HBV patients

with first- and second-generation nucleo(s)tide analogue treatment. We used a linear mixed model adjusted for preexisting comorbidities, geographic origin, and initial viral load to measure the individual variations of the eGFR computed with the Chronic Kidney Disease Epidemiology Collaboration formula. Although patients treated with ADV had a decreased eGFR over time, no renal impairment was detected with ETV and TDF treatment. Patients with preexisting CKD had a decreased eGFR, regardless of treatment. Patients without baseline CKD tended to have an improved eGFR over time, especially those with an initial high viral load born in an area of high HBV endemicity. Our results are in line with follow-up studies of registrational trials with ETV or TDF, indicating a creatinine level increase of more than 0.5 mg/dL in fewer than 1% of ETV- and TDF-treated patients after up to 5 and 3 years of ETV and TDF exposure, respectively.16,17 Reallife observational studies confirmed these trends: 418 and 302 ETV- and TDF-treated patients had a stable creatinine level after a median exposure of 52 and 33 months, respectively.18,19 In another study, serum creatinine level increases (
0.2 mg/dL) were common with both ETV and TDF, and were more frequent with ETV than with TDF.9 The problem is that CKD is a multifactorial disease; it therefore is difficult to draw any conclusion regarding the potential nephrotoxic or nephroprotective effect of a given drug. In 2 double-blind, placebo-controlled studies in patients with HBV infection and compensated liver disease, the prevalence of proteinuria, hematuria, and glycosuria was 17%, 30%, and 8%, respectively, in the placebo group.20,21 In the observational Hepatitis And Renal Parameters Evaluation study, the prevalence of proteinuria, hematuria, glycosuria, and uninfectious leukocyturia in untreated chronic HBV carriers was 38.1%, 20.6%, 3.9%, and 9%, respectively. Diabetes, hypertension, and dyslipidemia were observed, respectively, in 4.6%, 9.2%, and 38.8% of these patients.22 Renal evaluations in the Virgil European study

FLA 5.2.0 DTD
YJCGH54078_proof
5 December 2014
3:27 pm
ce DVC

Q18

523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580

6

581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638

Q19

Q20

Mallet et al

Clinical Gastroenterology and Hepatology Vol.

(n ¼ 381 treated patients) and in a monocentric study (n ¼ 264, including 132 NUC-treated and 132 controls) found that 1 of 10 to 1 of 5 of treated patients had renal deterioration with treatment (12% of treated as compared with 7% of untreated patients; P ¼ .05), which was observed more frequently with past or ongoing ADV treatment.23 On the opposite, another study compared 2 groups of HBV-infected age- and sexmatched patients treated with TDF or ETV (82 patients each) with a median duration of 2.7 and 4.1 years, respectively. In multivariate analysis, the 2 factors associated with renal deterioration were diabetes (P ¼ .004) and renal or liver transplantation (P ¼ .002), but not NUC exposure. A creatinine level increase greater than 0.2 mg/dL was more frequent with ETV (11%) than with TDF (2%) (P ¼ .02), but dose modifications were statistically different (5% vs 17%, respectively; P ¼ .0004).24 To sum up, all of these results are controversial and it is noteworthy that most of these real-life studies have not been published as full reports. The various results reported in these studies, including ours, probably are related to methodology. Most studies of kidney tolerance of NUCs have methodologic issues, including the use of inappropriate formulas to compute eGFR, the insensitive comparison of mean changes in eGFRs/creatinine levels or of number of patients with an increase in creatinine level over an arbitrary threshold, the absence of integration of baseline risk factors of CKD, or the use of a statistical methodology that does not take into account repeated measures. With a more stringent and sensitive approach, our conclusion is that renal function remains stable or improves with ETV and TDF in patients with normal baseline renal function. Active HBV replication is associated with glomerular lesions.1,25 We speculate that the observed increase in eGFR in the subgroup of patients with an initial high viral load is caused by HBV suppression, a result that we have reported with ADV in kidney transplant recipients and that others have reported with LAM or LdT.7,26,27 It is noteworthy that a German observational study found that ETV and TDF both are renoprotective at the individual level, with a methodology similar to that in the present work.8 This work did not take into account preexisting CKD before treatment, initial viral load, and ethnicity. We go further by showing that patients with a high viral load at baseline are more prone to increased eGFR with viral suppression and that people born in an area of high HBV endemicity are more subject to this effect. This is in line with analyses derived from the Globe study showing that patients treated with LdT have an improved eGFR with time.7 It is noteworthy that most trials of LdT included patients of Asian origin. Whether improvement of eGFR described in studies of LdT is caused by viral suppression, viral genotype, methodology, ethnicity, or by LdT itself remains to be determined. We also observed an increase in eGFR in patients with baseline cirrhosis who had an increased platelet level

-,

No.

-

with treatment; this could be owing to regression of cirrhosis.28 Finally, It has been suggested that a creatinine level increase or eGFR decrease are “late” markers of renal dysfunction and that “early” markers have to be implemented to diagnose early proximal tubular dysfunction. With such markers (glycosuria in the absence of diabetes, hyperuricemia, increased a1microglobulin/creatinine ratio, hypophosphoremia, phosphorus renal loss by the TmPO4/GFR ratio, which evaluates the tubular reabsorption of phosphorus), tubular dysfunction may be diagnosed in approximately one fourth of TDF-treated patients, increasing to 40% in patients with baseline risk factors of CKD. A limitation of our study was that it did not evaluate proximal tubular function. However, it is known from preclinical studies that if nucleotides have a competitive tubular toxicity on tubular cells, the toxicity is observed mainly with cidofovir, to a lesser extent with ADV, and is very limited with TDF.5 Even if tubular dysfunction was reported in first-line therapies with TDF in HIV-infected patients, including Fanconi syndrome, only 2 cases have been published in HBV monoinfected patients.29 In our series, serum phosphorus monitoring was performed regularly and there was no evidence of tubular dysfunction. The long-term impact, if any, of asymptomatic tubular dysfunction seems to be poor, at least on the basis of dual-energy x-ray absorptiometry scan studies on the bone mineral density loss in patients treated with TDF.30 However, given the rather short follow-up period of our study, we cannot exclude a very long-term toxic effect of NUCs in real life. Finally, we would like to point out that tubular toxicity is mainly dose-dependent and may require, in patients with baseline renal comorbidities, serum drug monitoring to avoid drug overdose and secondary renal impairment, which is mainly reversible with drug discontinuation or dosing adjustment. In conclusion, our study provides further evidence that second-generation NUC therapy, including TDF, does not alter renal function with time in patients with chronic HBV when risk factors of CKD are taken into account. NUC therapy may ameliorate infraclinical HBVrelated nephropathy in subgroups of patients. Whether NUC treatment diminishes HBV burden of disease in terms of CKD, especially in aging populations, has yet to be determined.

Uncited References This section consists of references that are included in the reference list but are not cited in the article text. Please either cite each of these references in the text or, alternatively, delete it from the reference list. If you do not provide further instruction for this reference, we will retain it in its current form and publish it as an “un-cited reference” with your article.31,32

FLA 5.2.0 DTD
YJCGH54078_proof
5 December 2014
3:27 pm
ce DVC

Q21

Q22

Q23

Q24

Q25

Q26

Q27

639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696

-

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

2014

Renal Effect of NUC in Chronic Hepatitis B

Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at www.cghjournal.org, and at http://dx.doi.org/10.1016/j.cgh.2014.11.021. Q28

References 1. Combes B, Shorey J, Barrera A, et al. Glomerulonephritis with deposition of Australia antigen-antibody complexes in glomerular basement membrane. Lancet 1971;2:234–237. 2. Knecht GL, Chisari FV. Reversibility of hepatitis B virus-induced glomerulonephritis and chronic active hepatitis after spontaneous clearance of serum hepatitis B surface antigen. Gastroenterology 1978;75:1152–1156. 3. Conjeevaram HS, Hoofnagle JH, Austin HA, et al. Long-term outcome of hepatitis B virus-related glomerulonephritis after therapy with interferon alfa. Gastroenterology 1995;109:540–546. 4. EASL Clinical Practice Guidelines. Management of chronic hepatitis B virus infection. J Hepatol 2012;57:167–185. 5. Fontana RJ. Side effects of long-term oral antiviral therapy for hepatitis B. Hepatology 2009;49:S185–S195. 6. Izzedine H, Launay-Vacher V, Isnard-Bagnis C, et al. Antiviral drug-induced kidney and electrolytes disorders. Minerva Urol Nefrol 2003;55:157–172. 7. Gane EJ, Deray G, Liaw YF, et al. Telbivudine improves renal function in patients with chronic hepatitis B. Gastroenterology 2014;146:138–146 e5. 8. Mauss S, Berger F, Filmann N, et al. Effect of HBV polymerase inhibitors on renal function in patients with chronic hepatitis B. J Hepatol 2011;55:1235–1240. 9. Gish RG, Clark MD, Kane SD, et al. Similar risk of renal events among patients treated with tenofovir or entecavir for chronic hepatitis B. Clin Gastroenterol Hepatol 2012;10: 941–946; quiz e68. 10. Lampertico P, Vigano M, Yurdaydin C, et al. Effectiveness and safety of tenofovir disoproxil fumarate in field practice: a multicenter European cohort study of 737 patients with chronic hepatitis B. Hepatology 2010;52:503A-A. 11. van Bommel F, de Man R, Ferenci P, et al. Evaluation of safety and tolerability of tenofovir DF (TDF) long term monotherapy in nucleos(t)ide analogue experienced patients with HBV monoinfection. J Hepatol 2009;50:S338-S. 12. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373–383. 13. Stevens PE, Levin A. Evaluation and management of chronic kidney disease: synopsis of the kidney disease: improving global outcomes 2012 clinical practice guideline. Ann Intern Med 2013;158:825–830. 14. Massy ZA, Khoa TN, Lacour B, et al. Dyslipidaemia and the progression of renal disease in chronic renal failure patients. Nephrol Dial Transplant 1999;14:2392–2406. 15. Qaseem A, Hopkins JRH, Sweet DE, et al. Screening, monitoring, and treatment of stage 1 to 3 chronic kidney disease: a clinical practice guideline from the American College of Physicians. Ann Intern Med 2013;159:835–847. 16. Chang TT, Lai CL, Kew Yoon S, et al. Entecavir treatment for up to 5 years in patients with hepatitis B e antigen-positive chronic hepatitis B. Hepatology 2010;51:422–430.

7

17. Heathcote EJ, Marcellin P, Buti M, et al. Three-year efficacy and safety of tenofovir disoproxil fumarate treatment for chronic hepatitis B. Gastroenterology 2011;140:132–143. 18. Lampertico P, Soffredini R, Vigano M, et al. Entecavir treatment for NUC naive, field practice patients with chronic hepatitis B: excellent viral suppression and safety profile over 5 years of treatment. Hepatology 2012;56:370A–371A. 19. Lampertico P, Soffredini R, Vigano M, et al. Tenofovir monotherapy suppressed viral suppression in most field practice, treatment-naive patients with chronic hepatitis B followed for 3 years in a multicenter European study. Hepatology 2012;56: 389A–A. 20. Hadziyannis SJ, Tassopoulos NC, Heathcote EJ, et al. Adefovir dipivoxil for the treatment of hepatitis B e antigen-negative chronic hepatitis B. N Engl J Med 2003;348:800–807. 21. Marcellin P, Chang TT, Lim SG, et al. Adefovir dipivoxil for the treatment of hepatitis B e antigen-positive chronic hepatitis B. N Engl J Med 2003;348:808–816. 22. Amet S, Bronowicki JP, Thabut D, et al. HARPE study: prevalence of renal abnormalities in chronic HBV infection. Hepatology 2012;56:638A–639A. 23. Deterding K, Lampe N, Reijnders J, et al. Prevalence and severity of kidney dysfunction in patients with chronic hepatitis B in Europe: data from the European Virgil Cohort. J Hepatol 2011;54:S147-S. 24. Gish RG, Mangahas MF, Baqai SF, et al. Risk of renal toxicity with tenofovir DF (TDF) for chronic hepatitis B (CHB). Hepatology 2010;52:529A-A. 25. Lai KN, Li PK, Lui SF, et al. Membranous nephropathy related to hepatitis B virus in adults. N Engl J Med 1991;324:1457–1463. 26. Fontaine H, Vallet-Pichard A, Chaix ML, et al. Efficacy and safety of adefovir dipivoxil in kidney recipients, hemodialysis patients, and patients with renal insufficiency. Transplantation 2005; 80:1086–1092. 27. Okuse C, Yotsuyanagi H, Yamada N, et al. Successful treatment of hepatitis B virus-associated membranous nephropathy with lamivudine. Clin Nephrol 2006;65:53–56. 28. Marcellin P, Gane E, Buti M, et al. Regression of cirrhosis during treatment with tenofovir disoproxil fumarate for chronic hepatitis B: a 5-year open-label follow-up study. Lancet 2013;381: 468–475. 29. Gracey DM, Snelling P, McKenzie P, et al. Tenofovir-associated Fanconi syndrome in patients with chronic hepatitis B monoinfection. Antivir Ther 2013;18:945–948. 30. Vigano M, Lampertico P, Soffredini R, et al. The course of bone mineral density in patients with chronic hepatitis B long-term treated with nucleos(t)ide analogs: a longitudinal cohort study. Hepatology 2011;54:1016A-A. 31. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150: 604–612. 32. Amet S, Bronowicki JP, Thabut D, et al. Prevalence of renal abnormalities in chronic HBV infection: the HARPE study. Liver Int 2014 Epub ahead of print.

Reprint requests Address requests for reprints to: Vincent Mallet, MD, PhD, Hépatologie, Assistance Publique-Hôpitaux de Paris, Hôpital Cochin, 27 Rue du Faubourg Saint Jacques, 75014 Paris, France. e-mail: [email protected]; fax: (xx) xxx-xxxx.

FLA 5.2.0 DTD
YJCGH54078_proof
5 December 2014
3:27 pm
ce DVC

755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 Q3Q4 811 812

8

813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870

Q5

Mallet et al

Clinical Gastroenterology and Hepatology Vol.

Acknowledgments The authors thank Drs Yacia Benai and Raoudha Akremi for supporting this work, Dr Amel Sahli for her statistical advice, and Dr Philippe Sultanik for fruitful discussions. Reproducible research statement: study protocol, statistical code, and data set are available from Vincent Mallet after establishing written agreement with the authors: e-mail: [email protected]. Conflicts of interest These authors disclose the following: Stanislas Pol has served as a speaker for GSK, BMS, Boehringer Ingelheim, Janssen, Gilead, Roche, MSD, Sanofi,

-,

No.

-

Novartis, Vertex, and AbbVie, has received grants from BMS, Gilead, Roche, and MSD, and is a board member of GSK, BMS, Boehringer Ingelheim, Janssen, Gilead, Roche, MSD, Sanofi, Novartis, Vertex, and AbbVie; Michael Schwarzinger has received a research grant from Novartis; Philippe Sogni has served as a speaker for BMS, Gilead, Roche, Janssen, and Merck and Co, and has served as a board member of Gilead, Merck and Co, and Janssen; and Vincent Mallet has served as a speaker for BMS, Gilead, Roche, Janssen, and Merck and Co, and has served as a board member of Gilead, Merck and Co, Q6 and Janssen. The remaining authors disclose no conflicts. Funding The Besafe study was funded with an unrestricted grant from Bristol-Myers Q7 Squibb.

FLA 5.2.0 DTD
YJCGH54078_proof
5 December 2014
3:27 pm
ce DVC

871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928

-

929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991

2014

Renal Effect of NUC in Chronic Hepatitis B

Appendix Figure 1. Flow chart. The Charlson comorbidity index predicts the 10-year mortality rate for a patient who may have a range of comorbid conditions, such as heart disease, acquired immune deficiency syndrome, or cancer (a total of 22 conditions). Each condition was assigned a score of 1, 2, 3, or 6, depending on the risk of dying associated with each condition. ESLD, end-stage liver disease; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HDV, hepatitis delta virus; SOT, solid-organ transplantation.

FLA 5.2.0 DTD
YJCGH54078_proof
5 December 2014
3:27 pm
ce DVC

8.e1 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 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