Value of Surveillance Studies for Patients With Stage I to II Diffuse Large B-Cell Lymphoma in the Rituximab Era

International Journal of

Radiation Oncology biology

physics

www.redjournal.org

Clinical Investigation

Value of Surveillance Studies for Patients With Stage I to II Diffuse Large B-Cell Lymphoma in the Rituximab Era Susan M. Hiniker, MD,* Erqi L. Pollom, MD,* Michael S. Khodadoust, MD, PhD,y Margaret M. Kozak, BA,* Guofan Xu, MD, PhD,z Andrew Quon, MD,z Ranjana H. Advani, MD,y and Richard T. Hoppe, MD* *Department of Radiation Oncology, yDivision of Oncology, Department of Medicine, and zDivision of Nuclear Medicine, Department of Radiology, Stanford Cancer Institute, Stanford, California Received Dec 17, 2014, and in revised form Jan 19, 2015. Accepted for publication Jan 27, 2015.

Summary Surveillance studies after the treatment of limited-stage diffuse large B-cell lymphoma (DLBCL) are widely used to monitor disease status, but with little evidence supporting their utility in improving outcome. We report our experience in a large cohort of patients treated for limited-stage DLBCL in the rituximab era. Our results demonstrate that although surveillance studies including positron emission tomography and computed tomography are frequently used after the completion of treatment and are associated with earlier detection of

Background: The role of surveillance studies in limited-stage diffuse large B-cell lymphoma (DLBCL) in the rituximab era has not been well defined. We sought to evaluate the use of imaging (computed tomography [CT] and positron emission tomography [PET]-CT) scans and lactate dehydrogenase (LDH) in surveillance of patients with stage I to II DLBCL. Methods: A retrospective analysis was performed of patients who received definitive treatment between 2000 and 2013. Results: One hundred sixty-two consecutive patients with stage I to II DLBCL were treated with chemotherapy þ/ rituximab, radiation, or combined modality therapy. The 5-year rates of overall survival (OS) and freedom from progression (FFP) were 81.2% and 80.8%, respectively. Of the 162 patients, 124 (77%) were followed up with at least 1 surveillance PET scan beyond end-of-treatment scans; of those, 94 of 124 (76%) achieved a complete metabolic response on PET scan after completion of chemotherapy, and this was associated with superior FFP (PZ.01, HRZ0.3) and OS (PZ.01, HR 0.3). Eighteen patients experienced relapse after initial response to therapy. Nine relapses were initially suspected by surveillance imaging studies (8 PET, 1 CT), and 9 were suspected clinically (5 by patient-reported symptoms and 4 by symptoms and physical examination). No relapses were detected by surveillance LDH. The median duration from initiation of treatment to relapse was 14.3 months among patients with relapses suspected by imaging, and 59.8 months among patients with relapses suspected clinically (PZ.077). There was no significant difference in OS from date of first therapy or OS after relapse between patients whose

Reprint requests to: Richard T. Hoppe, MD, Department of Radiation Oncology, Stanford Cancer institute, 875 Blake Wilbur Drive, Stanford, CA 94305-5847. Tel: (650) 723-6195; E-mail: [email protected] Int J Radiation Oncol Biol Phys, Vol. 92, No. 1, pp. 99e106, 2015 0360-3016/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ijrobp.2015.01.039

Presented in part at the 50th Annual Meeting of the American Society for Clinical Oncology, Chicago, IL, May 30 to June 3, 2014. Conflict of interest: none.

100

International Journal of Radiation Oncology  Biology  Physics

Hiniker et al.

relapse, there is no associated survival advantage.

relapse was suspected by imaging versus clinically. Thirteen of 18 patients underwent successful salvage therapy after relapse. Conclusions: A complete response on PET scan immediately after initial chemotherapy is associated with superior FFP and OS in stage I to II DLBCL. The use of PET scans as posttreatment surveillance is not associated with a survival advantage. LDH is not a sensitive marker for relapse. Our results argue for limiting the use of posttreatment surveillance in patients with limited-stage DLBCL. Ó 2015 Elsevier Inc. All rights reserved.

Introduction

Methods and Materials

The optimal treatment of patients with diffuse large B cell lymphoma (DLBCL) after successful initial therapy remains controversial. Given the possibility that curative salvage therapy may be more effective with low volume of disease, surveillance studies have been used frequently in patients after the completion of treatment (1). However, multiple studies, primarily among patients with advanced disease, have failed to show a clinical benefit of routine surveillance imaging at regular intervals (2-7), leading to recent clinical recommendations discouraging their use (8). Although the majority of these reports have not found a clinical benefit of surveillance imaging, others suggest that earlier detection of relapse may be associated with improved survival in a subset of patients (9, 10). The current guidelines of the National Comprehensive Cancer Network advise surveillance computed tomographic (CT) scans no more frequently than every 6 months for the first 2 years after treatment as clinically indicated (11). Many of the studies regarding surveillance imaging include a broad spectrum of patients, including patients treated in the pre-rituximab era, with variable inclusion criteria including stage and International Prognostic Index (IPI) score. Importantly, these studies include limited data on patients who present initially with limited disease. Approximately 30% of patients with DLBCL present with limited-stage disease (stage I-II). The prognosis of these patients treated in the rituximab era is excellent, with 3-year survival rates approaching 90% (12, 13). Even patients who experience relapse can frequently be treated successfully with curative salvage therapy (14). This population of patients with DLBCL with limited-stage disease and higher probability of curative salvage therapy represent a special population in whom earlier detection of relapse may be of particular benefit. We sought to analyze the role of surveillance imaging in the treatment of patients with limited-stage DLBCL. The objective of this study was to investigate the patterns of follow-up studies in the surveillance of patients with limited-stage DLBCL in the rituximab era, to assess the association with earlier detection of relapse, and to assess the potential association with clinical outcomes. We report our experience with 162 patients with stage I to II DLBCL treated in the rituximab era.

Patients We conducted a retrospective review of patients with newly diagnosed, stage I to II DLBCL treated with chemotherapy, radiation therapy, or both at our institution during the rituximab era, defined as during and after the year 2000. This study was approved by the institutional review board. The study was performed in accordance to the principles of the Declaration of Helsinki. From institutional databases, we identified 162 consecutive patients with stage I or II DLBCL treated during this time period. Eligible patients included those with histologically proven DLBCL, stage I or II according to the Ann Arbor staging system, including patients with bulky disease. The diagnosis of DLBCL was confirmed in every case in the laboratory of Surgical Pathology at our institution. We excluded patients with primary mediastinal lymphoma, primary central nervous system lymphoma, or transformed follicular lymphoma. Patient treatment was at the discretion of the treating physician. We collected and analyzed patient and treatment characteristics, including gender, age, stage, Eastern Cooperative Oncology Group (ECOG) performance status, initial lactate dehydrogenase (LDH), presence and site(s) of extranodal disease, size of disease and presence of bulky disease, stage-modified IPI score, chemotherapy and radiation therapy details, and follow-up information.

Patient follow-up After treatment, patients were reevaluated during regularly scheduled clinical visits. At our institution, patients are routinely followed up in the clinic every 3 months for the first 2 years of follow-up, then every 4 to 6 months until year 5, with annual visits afterwards. End-of-treatment scans were defined as the first scan to occur after completion of treatment, generally within 3 months from the end of therapy. Surveillance scans were defined as any routinely scheduled scans occurring after end-of-treatment scans, not prompted by clinical signs or symptoms. Surveillance imaging, including CT, positron emission tomography (PET), and magnetic resonance imaging, was performed at the discretion of the treating physician. Evaluation of treatment

Volume 92  Number 1  2015

response was assessed by clinical examination, radiographic studies, and laboratory studies including LDH. Response rates, which included imaging response to treatment, were classified as complete response (CR), partial response (PR), stable disease (SD), or progressive disease (PD) according to the 2007 International Working Group criteria (8). Local relapse was defined as disease occurring within the originally involved nodal area, extranodal areas, or both, or the radiation treatment field when radiation was administered as part of therapy.

Statistical analysis Time to event was measured from date of initiation of therapy. The Kaplan-Meier method was used to calculate estimates of overall survival (OS) and freedom from progression (FFP). A log-rank test was used to compare differences in Kaplan-Meier estimates between groups. Cox proportional hazard regression was used for univariate and multivariate analyses. A backward stepwise elimination approach, in addition to clinical judgment, was used for variable selection to construct the most parsimonious multivariate model possible. Only the patients who had assessment of metabolic response after chemotherapy (nZ124) were included in the final multivariate analyses. Patients who had progressive disease on postchemotherapy PET (nZ6) were excluded from the multivariate analysis for FFP because we were most interested in determining whether metabolic response after chemotherapy could predict for subsequent progression. Given the correlation between stage-modified IPI (SM-IPI) score and its component factors, we also looked at the individual SM-IPI component factors in a separate multivariable model without SM-IPI. The proportional hazards assumption of our final multivariate models was evaluated with Schoenfeld residuals and was met. The positive predictive value (PPV) and the negative predictive value (NPV) of metabolic response after chemotherapy for subsequent progression were calculated from the results in patients with progression after postchemotherapy PET surveillance scan. The normal approximation to the binomial distribution was used to calculate the 95% confidence interval (CI) for the NPV or PPV. All statistical tests were 2-sided, and a P value <.05 was considered significant. Survival analysis was performed by the R software (version 3.0.1), using the “survival” and “rms” packages.

Results Patient characteristics We identified 162 patients treated from 2000 and beyond with a median follow-up time of 56 months. The patient and treatment characteristics are detailed in Table 1. The median age was 60 years (range, 25-92 years). Staging

Surveillance studies in stage I-II DLBCL Table 1

101

Patient and treatment characteristics (nZ162) Characteristic

Gender Female Male Age, y, median 60 (range, 25-92) >60 60 Stage I II ECOG PS 0-1 >1 LDH Normal Elevated Extranodal disease None Present GI Bone Thyroid Breast Paranasal sinus Other (Waldeyer’s ring, testes etc.) Bulky disease (>7.5 cm) Yes No Stage-modified IPI score 0 or 1 >1 Chemotherapy regimen CHOP R-CHOP R-CEOP R þ other chemotherapy Therapy modality Chemotherapy alone Radiation therapy alone Combined chemotherapy and radiation

n (%) 60 (37) 102 (63) 82 (51) 80 (49) 81 (50) 81 (50) 136 (84) 26 (16) 113 (70) 49 (30) 64 98 24 22 9 8 8 27

(40) (61) (15) (14) (6) (4) (4) (17)

17 (10) 145 (90) 95 (59) 67 (41) 16 110 12 19

(10) (68) (7) (12)

42 (26) 2 (1) 118 (73)

Abbreviations: ECOG Z Eastern Cooperative Oncology Group; GE Z gastrointestinal; IPI Z International Prognostic Index; LDH Z lactate dehydrogenase; PS Z performance status; RZrituximab; R-CHOP Z rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone; R-CEOP Z rituximab, cyclophosphamide, etoposide, vincristine, and prednisone.

showed that 50% had stage I disease and 50% had stage II disease, with 17 patients (10%) having bulky disease defined as 7.5 cm in the greatest dimension. Adverse features were seen in a significant number of patients, including ECOG performance status of 2 (16%) and elevated LDH (30%). Extranodal disease was present in 98 patients (61%), the most common sites of extranodal disease being the gastrointestinal system (24 patients) and bone (22 patients). Ninety-five patients (59%) had SM-IPI scores of 0 or 1 at diagnosis, 67 patients (41%) had scores >1, and 28 patients (16%) had SM-IPI scores of 3 or 4 (Table 1).

International Journal of Radiation Oncology  Biology  Physics

Treatment details One hundred sixty patients (99%) received chemotherapy as part of their treatment regimen, and 2 patients (1%) received radiation therapy alone. One hundred ten patients (68%) were treated with at least 1 cycle of R-CHOP chemotherapy (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone). The median number of chemotherapy cycles received was 6. Owing to comorbidity or prior anthracycline, 12 patients (7%) received R-CEOP (rituximab, cyclophosphamide, etoposide, vincristine, and prednisone). An additional 19 patients (12%) were treated with an alternative rituximab-containing chemotherapy regimen. Although we limited our study to patients treated during the rituximab era, defined by the year 2000 and afterward, several patients treated at the beginning of that time frame did not receive rituximab as part of initial therapy, including 16 patients (10%) treated with CHOP chemotherapy without rituximab. One hundred eighteen patients (73%) received radiation therapy as consolidation after chemotherapy. Radiation therapy was individualized based on the site and extent of disease, but it generally conformed to involved field guidelines as determined by the treating radiation oncologist, and the median dose was 36 Gy.

Initial response to therapy Fourteen patients (9%) had primary refractory disease. Of 124 patients receiving a posttreatment PET scan, 94

Predictors for FFP and OS The 5-year rates of FFP and OS were 80.8% (95% CI, 74.4%87.8%) and 81.2% (95% CI 74.7%-88.3%), respectively (Fig. 1). Complete metabolic response on PET after completion of chemotherapy (PET CR) was associated with superior FFP (log-rank PZ.01) compared with those patients who achieved only PR or SD, and OS (log-rank P<.001) compared with those who achieved PR, SD, or PD (Fig. 2). We also examined other predictors for FFP and OS, including patient and treatment characteristics. We found that the following factors were also significant for FFP and OS on univariate analysis: age >60, PS  2, 2 or more sites of extranodal disease, SM-IPI 2 versus 1, and PET CR (Table 2). On multivariate analysis, only PET CR was significant for FFP (PZ.01; HRZ0.3; 95% CI, 0.2%-0.9%), and the presence of 2 or more sites of extranodal disease (PZ.02; HRZ6.7; 95% CI, 1.3-34.4), SM-IPI (PZ.005; HRZ5.4; 95% CI, 1.7%-17.5%), and PET CR (PZ.01; HRZ0.3; 95% CI, 0.1%-0.7%) were

Freedom from Progression

0.6

0.2

0.4 0.2

Overall Survival

0.6

0.8

1.0

B

0.8

A

patients (76%) had a metabolic CR, 22 patients (18%) had a PR, 2 patients (2%) had SD, and 6 patients (5%) had PD. Of all patients who had evaluation of metabolic response after chemotherapy, 20 patients ultimately experienced progression. Of the 94 patients who had a metabolic CR, 12 ultimately progressed (NPV 90%; 95% CI, 84.7%-95.3%). Of the remaining 24 patients who had either PR or SD, 8 ultimately experienced progression (PPV 56%; 95% CI, 47.3%-64.7%).

1.0

Hiniker et al.

0.4

102

0

143

1

Number at risk

121

101

2

3

81

4

72

5

58

6

40

25

7

Time from Treatment Start (Years)

8

19

9

13

10

0.0

0.0

Number at risk 162

162

0

130

109

91

74

63

52

36

23

19

1

2

3

4

5

6

7

8

9

13

10

Time from Treatment Start (Years)

Fig. 1. Overall survival (left) and freedom from progression (right) in limited-stage diffuse large B cell lymphoma in the rituximab era. Shaded area is 95% confidence interval.

Volume 92  Number 1  2015

103

1.0

1.0

Surveillance studies in stage I-II DLBCL

PETCR 0.8 0.6 0.2 Number at risk

Number at risk

30

17

11

6

1

94

74

47

34

14

0

2

4

6

8

1 noPETCR

0.0

0.0

no PETCR

0.4

Freedom from Progression

0.6 0.4

no PETCR

0.2

Overall Survival

0.8

PETCR

6 PETCR

10

24

14

10

7

2

94

68

44

30

13

0

Time from Treatment Start (Years)

2

4

6

2 noPETCR 6 PETCR

8

10

Time from Treatment Start (Years)

Fig. 2. Prognostic value of postchemotherapy complete response on overall survival and freedom from progression, by positron emission tomography. significant predictors of OS. When we examined the individual components of SM-IPI in a separate model without SM-IPI, PET CR remained the only significant predictor for FFP (PZ.02; HRZ0.3; 95% CI, 0.1%0.8%), and PET CR (PZ.002; HRZ0.2; 95% CI, 0.1%0.6%), PS  2 (PZ.03; HRZ2.9; 95% CI, 1.1%-7.5%), and presence of 2 or more sites of extranodal disease (PZ.007; HRZ6.3; 95% CI, 1.7%-23.6%) were significant predictors of OS.

imaging studies (8 PET, 1 CT), and 9 were suspected clinically (5 by patient-reported symptoms and 4 by both symptoms and physical examination). Patient-reported symptoms included worsening nasal obstruction and rhinitis in a patient with a nasal cavity/maxillary sinus recurrence, night sweats and abdominal discomfort in a patient with hepatic and splenic relapse, bone pain in a patient with relapsed bony disease, and neck swelling in 2 patients with cervical lymph node recurrence. The suspected sites of relapse were confirmed by biopsy before the institution of salvage therapy. No relapses were detected by surveillance LDH. At the time of relapse, LDH was checked in 13 patients and was found to be elevated in 3 patients, in all of whom the relapse was suspected clinically. The median duration from the initiation of treatment to relapse was 14.3 months (range, 7.8-121.1 months) among

Detection of relapse Eighteen patients ultimately experienced relapse after initial response to therapy. Of those, 17 (94%) had been followed up with surveillance imaging (Table 3). Nine of the 18 relapses were suspected initially by surveillance

Table 2

Univariate and multivariate analysis for FFP and OS Univariate FFP

Multivariate OS

FFP

OS

Characteristic

P

HR, 95% CI

P

HR, 95% CI

P

HR, 95% CI

P

HR, 95% CI

Age, y (>60) PS (>1) B symptoms (yes) Bulky (>7.5 cm) Elevated LDH (yes) Extranodal disease (yes) Extranodal disease (2 sites) SM-IPI (>1) PET CR after initial chemotherapy

.009 .1 .1 .08 .08 .17 .02 .007 .02

2.3, 1.2-4.3 2.0, 0.8-5.0 2.1, 0.8-5.0 2.0, 0.9-4.3 1.9, 0.9-3.8 0.6, 0.3-1.2 4.1, 1.2-14.0 2.7 1.3-5.4 0.3, 0.1-0.8

.001 <.0001 .4 .2 .06 .36 .002 <.0001 .007

4.1, 1.8-9.7 6.1, 2.9-12.9 1.5, 0.6-3.8 1.7, 0.8-3.8 2.0, 1.0-4.0 0.7, 0.4-1.5 5.4, 1.9-15.5 4.8, 2.2-10.5 0.2, 0.1-0.7

.2 .18 .01

3.2, 0.3-19.0 1.6, 0.7-4.6 0.3, 0.2-0.9

.02 .005 .01

6.7, 1.3-34.4 5.4, 1.7-17.5 0.3, 0.1-0.7

Abbreviations: CI Z confidence interval; FFP Z freedom from progression; HR Z hazard ratio; LDH Z lactate dehydrogenase; OS Z overall survival; PET CR Z positron emission tomography complete response; PS Z performance status; SM-IPI Z Stage-Modified International Prognostic Index.

International Journal of Radiation Oncology  Biology  Physics

Hiniker et al.

Management of relapses

Table 3 Patients undergoing surveillance: months after completion of treatment

There was no significant difference in survival from date of relapse (PZ.2) and from date of initial therapy (PZ.12) between patients whose relapse was suspected by imaging and those whose relapse was suspected clinically (Fig. 3, left). After relapse, 13 of 18 patients underwent successful salvage therapy, and 12 remain alive without second progression, with a median follow-up time of 15.6 months after initial relapse (Fig. 3, right). Salvage therapy included chemotherapy alone (10 patients), hematopoietic cell transplantation (4 patients), combined modality therapy (3 patients), and radiation therapy alone (1 patient). One patient declined additional therapy after relapse and remains alive with disease 3 months after relapse, and 1 patient experienced a second relapse 2.5 months after attaining a second remission and remains alive with disease 20.1 months after the second relapse. The remaining 4 patients died of disease or of treatment-related complications secondary to salvage therapy.

Months after completion of treatment: multivariate analysis Factor

3

6

9

12 18 24 >24

Patients undergoing 124 115 104 99 95 78 surveillance PET-CT Relapses detected by 1 4 1 0 1 1 surveillance Relapses detected clinically 0 2 1 0 1 1

45 0 4

Abbreviations: CT Z computed tomography; PET Z positron emission tomography.

patients with relapses suspected by imaging, and 59.8 months (range, 9.3-123.3 months) among patients with relapses suspected clinically (PZ.077). Eight of the 9 patients with relapse suspected by imaging experienced relapse within the first 3 years, and 4 of the 9 patients whose relapse was suspected clinically experienced relapse within the first 3 years. Remarkably, 5 patients had late relapses more than 5 years after the completion of therapy. There was nothing distinguishing about their initial presentation or treatment. All of these relapses were suspected clinically because these patients were no longer undergoing surveillance imaging during long-term follow-up. All had biopsy-confirmed DLBCL. Of the patients with imaging-suspected relapse, 4 had infield relapse alone, 2 had regional relapse with or without in-field relapse, and 3 had distant disease. Of the patients with relapse suspected clinically, 2 had regional relapse with or without in-field relapse, and 7 had distant disease. Clinical

1.0

A

Discussion In this retrospective study of DLBCL, we sought to clarify the role of posttreatment PET and surveillance PET imaging specifically in patients presenting with limited-stage disease. Our data suggest that PET-CT at the end of chemotherapy provides valuable prognostic information in limited-stage DLBCL. Complete metabolic response by posttreatment imaging was the only prognostic factor found to correlate with both FFP and OS on multivariate analysis. The strong prognostic impact of a negative PET-CT scan in limited-stage DLBCL at the end of therapy mirrors similar

B

1.0

104

0.8 0.6 0.4

Imaging

0.2

Freedom from Second Progression

0.6 0.4

Imaging

0.2

Overall Survival

0.8

Clinical

Number at risk 7

2 Clinical

9

6

4 Imaging

0

10 Time from Relapse (Months)

20

0.0

0.0

Number at risk 9

9

5

1 Clinical

9

4

3 Imaging

0

10

20

Time from Relapse (Months)

Fig. 3. Overall survival (left) and freedom from second progression (right) by detection method: imaging versus clinical signs and symptoms.

Volume 92  Number 1  2015

findings in nonestage stratified DLBCL in studies primarily composed of patients with advanced-stage disease (15-18). Our patients had excellent FFP and OS despite being a less favorable group of patients than those included in previous studies of limited-stage DLBCL (12). Specifically, we included patients with bulky disease, and a high proportion of patients presented with adverse prognostic factors including elevated LDH and poor ECOG performance status. The largest study to date of patients with all stages of DLBCL has shown no benefit of surveillance PET-CT imaging (7). In prospectively enrolled cohorts of patients at Mayo and in France, surveillance imaging detected relapse before clinical manifestations became evident in 1.6% and 1.8% of patients, respectively, but no difference was found in survival after relapse detected at scheduled follow-up versus before scheduled follow-up (7). Of note, approximately two thirds of the patients in each cohort had advanced-stage disease. In a separate study, Goldschmidt et al (19) also reported no survival advantage for patients whose relapse was found by scan before clinical symptoms became evident. However, others have noted that there may be subsets of the DLBCL population who benefit from surveillance imaging and that early detection of relapse may result in improved outcomes (9, 10). In certain settings, early detection of relapse may allow salvage therapies that would be not be feasible with a larger burden of disease (20). In this study, we sought to analyze whether patients with limited-stage DLBCL had better outcomes when relapses were detected by surveillance studies before the onset of clinical symptoms. The vast majority of patients in this study had at least 1 surveillance PET scan. We found a striking difference in the time to detection of relapse between patients with relapse detected by clinical findings compared with those detected by surveillance imaging (median, 59.8 months vs 14.3 months). This is at least in part the result of late relapses in several patients that occurred after surveillance imaging had been discontinued. Patients with early relapse (occurring within 3 years of treatment) did not have a significant difference in time to relapse detection between surveillance imaging versus clinical suspicion. Despite the difference in time to relapse detection, we did not find a survival advantage among patients with relapse detected by imaging in either early or total relapses. Patients with clinically suspected relapse had an excellent rate (89%) of successful salvage therapy, suggesting a limited benefit of early detection. The drawbacks to frequent scans are well reported and include financial costs per scan and radiation exposure, often in young patients, with the possible increase in risk of secondary malignancies (21). Additionally, patient anxiety has been shown to be associated with surveillance scans (22). In examining disease monitoring after completion of treatment, Hong et al (23) found an unacceptably high rate of false-positive surveillance imaging, frequently leading to unnecessary invasive biopsies in false-positive scans.

Surveillance studies in stage I-II DLBCL

105

Indeed, false-positive surveillance scans have been reported to be more frequent in patients treated with R-CHOP than in those treated with CHOP alone, with the effect lasting up to 3 years after treatment (5). In our cohort, we found that LDH was not a useful marker in the detection of relapse. Among the 18 patients who experienced relapse in our study, 3 of 18 had elevated LDH at the time of relapse, and in all of those the relapse was suspected clinically. LDH was not elevated in any asymptomatic patient whose relapse was suspected by surveillance imaging. As such, LDH does not appear to be a useful surveillance study for detecting relapse, in agreement with the results reported by other groups (24). The retrospective nature and small number of patients with relapsed disease in our study are notable limitations. Nevertheless, these results do not support the use of routine surveillance studies in limited-stage DLBCL, similar to other studies that have primarily been composed of patients with advanced-stage disease. Although surveillance imaging may result in earlier detection of relapse in limitedstage DLBCL, we find no benefit in terms of survival. Rather, appropriate follow-up and thorough clinical examinations may be sufficient for the identification of possible relapse, and patients with relapse have excellent rates of successful salvage and long-term survival. In conclusion, we report that surveillance imaging is frequently used after treatment of limited-stage DLBCL and is associated with earlier detection of relapse but no survival advantage. We believe that our data argue for reduced frequency of surveillance studies in patients with limited-stage DLBCL after the completion of treatment.

References 1. Abel GA, Vanderplas A, Rodriguez MA, et al. High rates of surveillance imaging for treated diffuse large B-cell lymphoma: Findings from a large national database. Leuk Lymphoma 2012;53:1113-1116. 2. Jerusalem G, Beguin Y, Fassotte MF, et al. Early detection of relapse by whole-body positron emission tomography in the follow-up of patients with Hodgkin’s disease. Ann Oncol 2003;14:123-130. 3. Zinzani PL, Stefoni V, Tani M, et al. Role of [18f]fluorodeoxyglucose positron emission tomography scan in the follow-up of lymphoma. J Clin Oncol 2009;27:1781-1787. 4. Cheah CY, Hofman MS, Dickinson M, et al. Limited role for surveillance PET-CT scanning in patients with diffuse large B-cell lymphoma in complete metabolic remission following primary therapy. Br J Cancer 2013;109:312-317. 5. Avivi I, Zilberlicht A, Dann EJ, et al. Strikingly high false positivity of surveillance FDG-PET/CT scanning among patients with diffuse large cell lymphoma in the rituximab era. Am J Hematol 2013;88:400-405. 6. Petrausch U, Samaras P, Haile SR, et al. Risk-adapted FDG-PET/CTbased follow-up in patients with diffuse large B-cell lymphoma after first-line therapy. Ann Oncol 2010;21:1694-1698. 7. Thompson CA, Ghesquieres H, Maurer MJ, et al. Utility of routine post-therapy surveillance imaging in diffuse large B-cell lymphoma. J Clin Oncol 2014;32:3506-3512. 8. Cheson BD, Fisher RI, Barrington SF, et al. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: The Lugano classification. J Clin Oncol 2014;32:3059-3068.

106

Hiniker et al.

9. Liedtke M, Hamlin PA, Moskowitz CH, et al. Surveillance imaging during remission identifies a group of patients with more favorable aggressive NHL at time of relapse: A retrospective analysis of a uniformly-treated patient population. Ann Oncol 2006;17:909-913. 10. El-Galaly TC, Mylam KJ, Bogsted M, et al. Role of routine imaging in detecting recurrent lymphoma: A review of 258 patients with relapsed aggressive non-Hodgkin and Hodgkin lymphoma. Am J Hematol 2014;89:575-580. 11. Zelenetz AD, Gordon LI, Wierda WG, et al. Non-Hodgkin’s lymphomas, version 2.2014. J Natl Compr Canc Netw 2014;12:916-946. 12. Persky DO, Unger JM, Spier CM, et al. Phase II study of rituximab plus three cycles of CHOP and involved-field radiotherapy for patients with limited-stage aggressive B-cell lymphoma: Southwest Oncology Group study 0014. J Clin Oncol 2008;26:2258-2263. 13. Pfreundschuh M, Trumper L, Osterborg A, et al. CHOP-like chemotherapy plus rituximab versus CHOP-like chemotherapy alone in young patients with good-prognosis diffuse large-B-cell lymphoma: A randomised controlled trial by the MabThera International Trial (MINT) Group. Lancet Oncol 2006;7:379-391. 14. Philip T, Guglielmi C, Hagenbeek A, et al. Autologous bone marrow transplantation as compared with salvage chemotherapy in relapses of chemotherapy-sensitive non-Hodgkin’s lymphoma. N Engl J Med 1995;333:1540-1545. 15. Pregno P, Chiappella A, Bello M, et al. Interim 18-FDG-PET/CT failed to predict the outcome in diffuse large B-cell lymphoma patients treated at the diagnosis with rituximab-CHOP. Blood 2012;119:2066-2073. 16. Spaepen K, Stroobants S, Dupont P, et al. Prognostic value of positron emission tomography (PET) with fluorine-18 fluorodeoxyglucose ([18f]FDG) after first-line chemotherapy in non-Hodgkin’s lymphoma: Is [18f]FDG-PET a valid alternative to conventional diagnostic methods? J Clin Oncol 2001;19:414-419.

International Journal of Radiation Oncology  Biology  Physics 17. Zijlstra JM, Lindauer-van der Werf G, Hoekstra OS, et al. 18F-fluorodeoxyglucose positron emission tomography for post-treatment evaluation of malignant lymphoma: A systematic review. Haematologica 2006;91:522-529. 18. Dabaja BS, Phan J, Mawlawi O, et al. Clinical implications of positron emission tomography-negative residual computed tomography masses after chemotherapy for diffuse large B-cell lymphoma. Leuk Lymphoma 2013;54:2631-2638. 19. Goldschmidt N, Or O, Klein M, et al. The role of routine imaging procedures in the detection of relapse of patients with Hodgkin lymphoma and aggressive non-Hodgkin lymphoma. Ann Hematol 2011; 90:165-171. 20. Lambert JR, Bomanji JB, Peggs KS, et al. Prognostic role of PET scanning before and after reduced-intensity allogeneic stem cell transplantation for lymphoma. Blood 2010;115:2763-2768. 21. Shenoy P, Sinha R, Tumeh JW, et al. Surveillance computed tomography scans for patients with lymphoma: Is the risk worth the benefits? Clin Lymphoma Myeloma Leuk 2010;10:270-277. 22. Thompson CA, Charlson ME, Schenkein E, et al. Surveillance CT scans are a source of anxiety and fear of recurrence in long-term lymphoma survivors. Ann Oncol 2010;21:2262-2266. 23. Hong J, Kim JH, Lee KH, et al. Symptom-oriented clinical detection versus routine imaging as a monitoring policy of relapse in patients with diffuse large B-cell lymphoma. Leuk Lymphoma 2014;55:23122318. 24. El-Sharkawi D, Basu S, Ocampo C, et al. Elevated lactate dehydrogenase levels detected during routine follow-up do not predict relapse in patients with diffuse large B-cell lymphoma who achieve complete remission after primary treatment with rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone-like immunochemotherapy. Leuk Lymphoma 2012;53:1949-1952.