Lymphocyte-Depleted Classical Hodgkin s Lymphoma: A Comprehensive Analysis From the German Hodgkin Study Group

Transcription

1 Published Ahead of Print on September 12, 211 as 1.12/JCO The latest version is at JOURNAL OF CLINICAL ONCOLOGY O R I G I N A L R E P O R T Lymphocyte-Depleted Classical Hodgkin s Lymphoma: A Comprehensive Analysis From the German Hodgkin Study Group Beate Klimm, Jeremy Franklin, Harald Stein, Dennis A. Eichenauer, Heinz Haverkamp, Volker Diehl, Michael Fuchs, Peter Borchmann, and Andreas Engert Beate Klimm, Dennis A. Eichenauer, Heinz Haverkamp, Volker Diehl, Michael Fuchs, Peter Borchmann, and Andreas Engert, German Hodgkin Study Group; Jeremy Franklin, University Hospital of Cologne, Cologne; and Harald Stein, Berlin Reference Center for Lymphoma and Hematopathology at the Pathodiagnostik Berlin, Berlin, Germany. Submitted April 11, 211; accepted July 14, 211; published online ahead of print at on September 12, 211. The German Hodgkin Study Group is supported by the Deutsche Krebshilfe and is part of the Kompetenznetz Maligne Lymphome supported by the Bundesministerium für Bildung und Forschung. Presented in part at 15th Congress of the European Hematology Association, June 1-13, 21, Barcelona, Spain. Authors disclosures of potential conflicts of interest and author contributions are found at the end of this article. Corresponding author: Andreas Engert, MD, Department I of Internal Medicine, University Hospital of Cologne, 5924 Cologne, Germany; a.engert@ uni-koeln.de. 211 by American Society of Clinical Oncology X/11/2999-1/$2. A B S T R A C T Purpose To investigate the clinical characteristics and treatment outcome of patients with lymphocytedepleted classical Hodgkin s lymphoma (CHL) compared with other histologic subtypes of Hodgkin s lymphoma (HL). Patients and Methods From a total of 12,155 evaluable patients with biopsy-proven HL treated within the German Hodgkin Study Group trials HD4 to HD15, 1,19 patients underwent central expert pathology review. Eighty-four patients with CHL ( 1%) were identified and confirmed. The median follow-up time was 67 months. Results Patients with CHL, compared with patients with other histologic subtypes, presented more often with advanced disease (74% v 42%, respectively; P.1) and B symptoms (76% v 41%, respectively; P.1). Other risk factors were also more frequent in patients with CHL. Complete remission or unconfirmed complete remission was achieved in 82% of patients with CHL compared with 93% of patients with other HL subtypes (P.1), and more patients with CHL had progressive disease. At 5 years, progression-free survival (PFS) and overall survival (OS) were significantly lower in patients with CHL compared with patients with other HL subtypes (PFS, 71% v 85%, respectively; P.1; OS, 83% v 92%, respectively; P.18). However, when analyzing the subgroup of patients who underwent treatment with intensified or dose-dense bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone, patients with CHL (n 39) had similar outcomes when compared with patients with other subtypes of HL (n 3,564; P.61). Conclusion CHL has a different pattern from other HL subtypes with more clinical risk factors at initial diagnosis and significantly poorer prognosis. Patients with CHL should be treated with modern dose-intense treatment strategies. J Clin Oncol by American Society of Clinical Oncology DOI: 1.12/JCO INTRODUCTION Lymphocyte-depleted classical Hodgkin s lymphoma (CHL) is a rare entity, accounting for less than 1% of all patients with newly diagnosed Hodgkin s lymphoma (HL) in Western countries. Because of its rarity, little is known about clinical characteristics, course, and treatment outcome of patients with CHL. The WHO classification generally distinguishes between lymphocytepredominant (LP) HL and classical HL, which has been further classified into the following four subtypes: lymphocyte rich classical (LRC), nodular sclerosing (NS), mixed cellularity (MC), and lymphocyte depleted (). 1 Recently, other less frequent subtypes such as LRC and LP were described in more detail. 2,3 Histologically, the following two variants were characterized: a diffuse fibrosis variant with abundant Hodgkin s/reed-sternberg cells in a hypocellular background showing disordered fibrosis, rich in histiocytes but with few lymphocytes; and a reticular variant, showing numerous Hodgkin s/ Reed-Sternberg cells with bizarre cytologic features (Appendix Fig A1, online only). Exact diagnosis requires immunohistochemical staining, which aids to distinguish CHL from other HL and non-hl (NHL) subtypes by American Society of Clinical Oncology 1

2 Klimm et al Those few patients with histology who are analyzed in case series or included in clinical trials generally seem to have a poorer prognosis than patients with other HL subtypes. Current multimodality treatment protocols that achieve better tumor control might have blunted the effect of histology and other historical risk factors on outcome. 7 In contrast, results from population-based analyses suggest that morphologic groups continue to have prognostic significance, even in the modern treatment era, with a significantly poorer prognosis of MC and subtypes. 8 However, no larger cohort of patients with CHL has been analyzed thus far, because numbers were too small for reliable conclusions. To shed more light on the clinical course of CHL, we revisited our database to analyze patient characteristics and treatment outcome. PATIENTS AND METHODS Patient Selection From a total of 12,155 evaluable patients with biopsy-proven HL treated within four generations of German Hodgkin Study Group (GHSG) trials (HD4 to HD15), 1,19 patients underwent central expert pathology review and were included in this analysis. 9-2 Eighty-four patients (.84%) with - CHL and 9,935 patients with other HL histology were identified by the review board. Data were taken from the latest final or follow-up analysis (HD4 to HD12) or from the latest interim analysis (HD13 to HD15). In most GHSG trials, patients were between 16 and 75 years of age. To be eligible for random assignment, they had to have biopsy-proven HL histology at diagnosis, a creatinine clearance more than 6 ml/min, serum aminotransferases less than 3 the upper normal limit, bilirubin less than 2 mg/dl, left ventricular ejection fraction more than.45, and forced expiratory volume in the first second or diffusion capacity of carbon monoxide more than 6% of predicted. Required blood cell count was defined as WBCs 3,5/ L, platelets 1,/ L, and hemoglobin level 8 g/dl. Randomly assigned patients had to be negative for antibodies against HIV and free of active infection. All patients signed informed consent before study entry, which was based on institutional review board guidelines. All protocols were approved by the ethics committee at each participating institution and were conducted in accordance with the Declaration of Helsinki. Treatment regimens are listed in Appendix Table A1 (online only). Staging Procedures and Treatment Histopathologic diagnosis was made initially by the local pathologist who then sent paraffin block biopsy samples for central pathology review to the GHSG expert pathology panel. Diagnostic criteria for identification of CHL were used as described by Pileri et al. 5 The extent of disease was assessed by chest x-ray, abdominal ultrasound, computed tomography, isotopic bone scan, and bone marrow biopsy. Patients with clinical stage I or II disease without risk factors, such as large mediastinal mass, extranodal lesions, massive spleen involvement, elevated erythrocyte sedimentation rate (ESR; 5 mm without B symptoms and 3 mm with B symptoms), and involvement of three or more lymph node areas, were included in studies for early favorable stages (HD4, HD7, HD1, and HD13). Patients with clinical stage I or IIA disease with one or more of these risk factors and stage IIB with certain risk factors only (B symptoms and/or three lymph node areas involved) were included in trials for early unfavorable stages (HD5, HD8, HD11, and HD14). Patients with clinical stage IIB disease with the risk factors of large mediastinal mass and/or extranodal involvement and patients with clinical stage III or IV were allocated to the advanced risk group and treated accordingly (HD6, HD9, HD12, and HD15). Between the different GHSG trial generations, staging criteria were slightly altered; for examples, massive spleen involvement was abolished as separate risk factor after the third generation, and age limits for patients in unfavorable and advanced stages receiving bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone (BEACOPP) chemotherapy were introduced (18 to 6 years). These smaller adjustments in terms of age, staging, and risk factor profile were implemented at the beginning of a new trial generation for all patients irrespective of histology. Random assignment strategies including the chemotherapeutic regimens and the amount of radiotherapy applied in GHSG trials HD4 to HD15 have been described elsewhere and are listed in Table A Response and relapse criteria also have been described elsewhere. 9-2 Statistical Methods Statistical analyses were performed with SAS Version 9.2 (SAS Institute, Cary, NC). Differences between histologic subtypes in the distribution of demographic and clinical variables were assessed using the 2 test or Fisher s exact test for dichotomous variables. Progression-free survival (PFS) and overall survival (OS) were estimated according to the Kaplan-Meier method 21 and compared between groups using the log-rank test. All reported P values are two-sided. All variables found to have a P.5 were considered significant. PFS was calculated from the date of random assignment; was defined as time until progression, relapse, or death from any cause; and was censored at the date of last information on tumor status. OS was defined from the date of random assignment until death as a result of any cause and was censored at the date of last information. Cox proportional hazards regression model was used to assess the influence of histology on PFS, 22 allowing for the influences of demographic and clinical variables. These were selected for inclusion in the model using a forward stepwise procedure. RESULTS Patient Characteristics In the pooled total of 1,19 patients, 84 patients with CHL were identified, accounting for.84% of all patients with HL. Interestingly, the proportion of patients with HL diagnosed from 1998 to 28 with subtype varied between.4% and 2% per year, with decreasing tendency. Almost all patients with CHL were diagnosed in early unfavorable or advanced stages; only one patient was diagnosed as early favorable. The numbers of patients with CHL per trial are listed in Table 1. Baseline patient demographics and clinical characteristics of patients with and other subtypes of HL were clearly different. Patients with histology, compared with patients with non- HL, presented more often with advanced disease (74% v 42%, respectively; P.1) and B symptoms (76% v 41%, respectively; P.1). Certain risk factors occurred more frequently in patients with CHL than in patients with other HL subtypes, such as large mediastinal mass (32% v 19%, respectively; P.46), extranodal involvement (25% v 14%, respectively; P.11), high ESR (76% v 47%, respectively; P.1), and involvement of three or more lymph node areas (88% v 6%, respectively; P.1). Furthermore, newly diagnosed patients with CHL, compared with other patients, more often showed involvement of bone marrow (11% v 4%, respectively; P.98) and liver (19% v 4%, respectively; P.1). Patients with CHL presented significantly more often with a higher International Prognostic Score (IPS; P.1) 23 compared with patients with non- HL (Table 2). Except for the GHSG risk factor of extranodal involvement (P.65), all described differences in baseline demographics and clinical characteristics remained consistent and significant when comparing patients with subtype directly with patients with NS or MC subtypes (Table 2). This ensured that the better risk profile of the patients with non- HL was not a result of patients with LRC or LP by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

3 Lymphocyte-Depleted Classical Hodgkin s Lymphoma Trial Table 1. Patients With CHL in Trials HD4 to HD15 Stage All Patients (No.) Patients With CHL (No.) HD4-HD6 1, HD4 Early favorable 375 HD5 Early unfavorable HD6 Advanced HD7-HD9 3,27 24 HD7 Early favorable 627 HD8 Early unfavorable 1,136 4 HD9 Advanced 1,264 2 HD1-HD12 4, HD1 Early favorable 1,176 1 HD11 Early unfavorable 1,394 9 HD12 Advanced 1, HD13-HD15 3,39 12 HD13 Early favorable 414 HD14 Early unfavorable 1,13 2 HD15 Advanced 1,612 1 Total 12, Abbreviations: HD, Hodgkin s disease; CHL, lymphocyte-depleted classical Hodgkin lymphoma. Of 12,155 eligible patients with biopsy-proven Hodgkin s lymphoma, 1,19 patients underwent central expert pathology review (84 patients with CHL and 9,935 patients with other subtypes of Hodgkin s lymphoma). subtypes. These patients have been shown to have a better prognosis than patients with NS or MC subtypes. 2,3 Response to Treatment and Causes of Death The overall response rates were as follows. Complete remission was recorded in significantly fewer patients with histology compared with patients with non- subtypes (82% v 93%, respectively; P.1). There was no further difference between NS (93%) and MC (94%) subtypes. Partial remission was documented in 4.8% of patients with HL and 1.3% of patients with non- HL, and primary progressive disease was documented in 4.8% of patients with HL and 3.1% of patients with non- HL. In total, 9% of all patients (n 933) died, including 2% of patients with, 8% with NS, and 1% with MC HL. The causes of death during the study and follow-up period included death from HL, toxicity from primary or salvage treatment or autologous transplantation, secondary malignancies, and more rare causes, such as concomitant disease with organ dysfunction, suicide, accident, and others. Deaths as a result of HL or treatment toxicity were more frequently recorded in patients with subtype (Table 3). Secondary malignancies including acute myeloid leukemia, NHL, and solid tumors were documented in 4.8% of patients with, 3.7% of patients with NS, and 6.3% of patients with MC HL (data not shown). Treatment Outcome Median observation time for PFS was 67 months. Patients with CHL showed a significantly inferior PFS compared with all patients with non- subtype (P.1) and patients with NS and MC subtypes (P.1; Fig 1). Five-year PFS rates were 71% for CHL (95% CI, 59.9% to 79.9%), 85% for non-chl (95% CI, 83.8% to 85.3%), 85% for NS (95% CI, 84.4% to 86.4%), and 84% for MC (95% CI, 82.% to 85.2%; P.1). Inferiority was also observed for Table 2. Baseline Patient Demographics and Clinical Characteristics Demographic or Clinical Characteristic Non- (n 9,935) Review Histology (n 84) NS (n 5,9) MC (n 2,494) No. % No. % No. % No. % Age, years , , , , , , Sex Female 4, , Male 5, , , Stage, GHSG Early favorable 2, Early unfavorable 3, , Advanced 4, , ,8 4.4 Clinical stage I 1, II 4, , III 2, , IV 1, B symptoms 4, , Risk factors, GHSG Large mediastinal mass 1, , Extranodal involvement 1, High ESR 4, , lymph node areas 5, , , Organ involvement Bone marrow Lung Liver Skeletal IPS risk factors 1, , , , , , Abbreviations: ESR, erythrocyte sedimentation rate; GHSG, German Hodgkin Study Group; IPS, International Prognostic Score;, lymphocyte depleted; MC, mixed cellularity; NS, nodular sclerosis. OS of patients with CHL compared with non-chl (P.18) or compared with NS or MC histology (P.12; Fig 2). Five-year OS rates were 83% for CHL (95% CI, 72.3% to 89.3%), 92% for non-chl (95% CI, 91.4% to 92.5%), 93% for NS (95% CI, 91.9% to 93.3%), and 91% for MC (95% CI, 89.9% to 92.4%; P.1). Furthermore, log-rank test analysis stratified by early unfavorable and advanced stages continued to show inferiority in terms of PFS (P.93) and OS (P.19) for patients with CHL compared with patients with non- subtypes. In similar stratified analyses comparing patients with CHL with patients with combined by American Society of Clinical Oncology 3

4 Klimm et al Table 3. Response to Treatment and Causes of Death Response to Treatment and Causes of Death Non- (n 9,935) Review Histology (n 84) NS (n 5,9) MC (n 2,494) No. % No. % No. % No. % Response CR CRu 9, , , CR 4, , , CRu 4, , , PR NC PD Unknown Causes of death Hodgkin s lymphoma Toxicity of primary treatment Toxicity of salvage treatment Treatment/ASCT Secondary malignancy Other or unknown Abbreviations: ASCT, autologous stem-cell transplantation; CR, complete remission; CRu, complete remission unconfirmed;, lymphocyte depleted; MC, mixed cellularity; NC, no change; NS, nodular sclerosis; PD, progressive disease; PR, partial remission. NS MC HL, the P values were.52 for PFS and.11 for OS (data not shown). Risk Factors Multivariate analyses were performed to assess the influence of histology on PFS. The variables of histology ( v NS MC), age, sex, stage, B symptoms, and IPS and the risk factors of large mediastinal mass, extranodal involvement, high ESR, and involvement of three lymph node areas were tested in a stepwise Cox regression model, including variables with P.5 and excluding variables with P.1. The optimal model included the variables of age (P.1), IPS (P.1), stage (P.1), three lymph node areas involved (P.19), large mediastinal mass (P.3), and high ESR (P.42). histology (P.12) was not significant based on this model. Even when the model was restricted to the generally recognized clinical risk factors of stage, age, sex, and IPS, no further prognostic significance was found for the factor of histology (P.16; data not shown). Role of Modern Chemotherapy In a Kaplan-Meier analysis restricted to the subgroup of patients (n 3,66) who underwent modern intensified or dose-dense treatment regimens, such as BEACOPP-escalated or BEACOPP-14 (ie, HD9 arm C, HD12, HD14 arm B, and HD15), patients with CHL (n 39) no longer had an inferior PFS compared with patients with non- subtypes (P.61; Fig 3). Five-year PFS rates were 83% (95% CI, 65.7% to 92.1%) for versus 87% (95% CI, 85.8% to 88.3%) for non-. OS was also not inferior (P.55). These results are important with regard to CHL in advanced stages because almost all patients with histology treated with BEACOPP-escalated or BEACOPP-14 belonged to that risk group (n 38). In the group of A Progression-Free Survival B Not P <.1 Not 9,935 7,23 4,759 2, Progression-Free Survival NS MC P < NS 5,9 4,371 2,865 1, MC 2,494 1,753 1, Fig 1. Progression-free survival (PFS) by histologic subtype: (A) lymphocyte depleted () versus non-, and (B) versus nodular sclerosis (NS) and mixed cellularity (MC). Nos. at risk are shown at, 3, 6, 9, and 12 months. Error bars indicate 95% CI. comparable patients (n 2,26) treated without BEACOPP-escalated or BEACOPP-14, patients with CHL had significantly worse 5-year PFS (62%) than patients with non- subtypes (84%; P.45); OS at 5 years was also inferior for patients with CHL (P.23) in this subgroup (data not shown). DISCUSSION Although the histologic and molecular features have been characterized in part before, 4-6,24 little is known about the clinical characteristics of CHL. The aim of this retrospective study was thus to describe the clinical features of CHL and to determine treatment outcome for this rare entity, accounting for.84% of all our patients with HL. We analyzed 1,19 patients with biopsy-proven HL from four generations of clinical trials registered in the GHSG database who had undergone central expert pathology review. The following findings emerge from this study. Compared with other HL entities, patients with CHL presented more often with unfavorable characteristics by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

5 Lymphocyte-Depleted Classical Hodgkin s Lymphoma A 1. A 1. Overall Survival Not P =.2 Progression-Free Survival Not P =.613 B Not 9,935 8,4 5,767 3, Overall Survival NS MC P < NS 5,9 4,859 3,414 1, MC 2,494 1,97 1, Fig 2. Overall survival (OS) by histologic subtype: (A) lymphocyte depleted () versus non-, and (B) versus nodular sclerosis (NS) and mixed cellularity (MC). Nos. at risk are shown at, 3, 6, 9, and 12 months. Error bars indicate 95% CI. such as advanced stage, B symptoms, large mediastinal mass, extranodal disease, high ESR, involvement of three or more lymph node areas, and higher IPS; in addition, organ involvement of bone marrow and liver was also more frequent. Patients with CHL, compared with patients with other HL subtypes, had significantly poorer rates for complete remission (82% v 93%, respectively; P.1), PFS (71% v 85%, respectively; P.1), and OS (83% v 92%, respectively; P.1). Inferiority was still significant after stratification of patients with CHL for early unfavorable or advanced stages according to the GHSG risk stratification. Multivariate analysis identified age (P.1), IPS (P.1), stage (P.1), three lymph node areas involved (P.19), large mediastinal mass (P.3), and high ESR (P.42) as adverse prognostic factors for PFS. Importantly, histology was not identified as an independent prognostic factor. Patients with CHL in advanced stages treated with BEACOPP-escalated or BEACOPP-14 had similar outcome compared with patients with advanced-stage non-chl, suggesting that these patients should be treated with dose-intense treatment strategies. The proportion of patients with newly diagnosed CHL in this analysis varied between 2% and.4% per year from 1988 to 28 and Not 3,564 2,353 1, B Progression-Free Survival Not P = Not 2,238 1,711 1, Fig 3. Progression-free survival (PFS) of patients with lymphocyte-depleted () versus non- histology in early unfavorable and advanced stages treated (A) with or (B) without bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone (BEACOPP) escalated or BEACOPP-14. Nos. at risk are shown at, 3, 6, 9, and 12 months. Error bars indicate 95% CI. decreased over time. The reasons are not fully understood. It is possible that the incidence of CHL is really slowly decreasing. In addition, most patients with CHL were diagnosed in advanced stages, and 1% were older than age 6 years. Because more recent GHSG trials for advanced stages (HD12 and HD15) excluded patients older than age 6 years, this might have decreased the proportion of patients with CHL documented in the study population. In addition, there has been a constant improvement in pathology over the last decades with more accurate definition of all HL subtypes. 4,5 Before 198, patients with aggressive NHL were sometimes misdiagnosed as having HL and classified as or MC subtypes. 25 In one report, this led up to 22% of all patients with HL being originally classified as CHL before adequate revision of diagnosis. 26 Today, expert pathologists more reliably distinguish CHL from other entities including diffuse large B-cell lymphoma, anaplastic large-cell lymphoma, or gray zone lymphoma by means of morphology, immunophenotyping, and molecular genetic analyses. 6 So far, only small case series evaluating clinical features and prognosis of patients with CHL have been published. An early by American Society of Clinical Oncology 5

6 Klimm et al report suggested a poorer prognosis for patients with CHL compared with patients with other HL subtypes. However, there was no difference in survival when patients with CHL were treated with single-agent chemotherapy in the 196s or multiple agent chemotherapy in the 197s. 27 In 1986, Kant et al 26 reviewed all patients at the National Cancer Institute who were classified as having CHL between 1964 and The subtype was confirmed in nine of the 43 patients originally diagnosed. In their series of patients with CHL with up-to-date diagnostics after adequate pathology review, the prognosis was not different from other histopathologic subtypes of HL. Greer et al 28 performed a clinicopathologic review of 25 patients with CHL, which is the largest case series published so far. The clinical presentation of CHL was similar to our series, including more patients with B symptoms (92%), subdiaphragmatic disease (88%), bone marrow involvement (56%), and advanced-stage disease (1%). Most patients had received mechlorethamine, vincristine, procarbazine, and prednisone chemotherapy at that time, and their prognosis was poor, with a median survival time of 36 months. Other reports suggested that negative prognostic factors such as subdiaphragmatic disease or tissue eosinophilia, which were more frequently observed in CHL, contributed to the poorer prognosis of these patients. 29,3 What are the clinical implications of the present analysis? It is demonstrated that histology is associated with poorer prognostic factors and poorer overall outcome. Importantly, patients with histology in advanced stages have a similar prognosis to other patients with HL when treated with modern dose-intense regimens, suggesting that the earlier impact of histology as prognostic marker has diminished with the development of more effective chemotherapy. Thus, one of the questions raised by this analysis is whether patients with CHL generally should receive more intensive treatment. Our results imply that patients with CHL should receive a dose-intense chemotherapy such as BEACOPP-escalated or BEACOPP-14. However, only one patient with CHL in early unfavorable stages was treated with BEACOPP-escalated (arm B of HD14), and only one patient with histology was diagnosed with an early favorable stage and subsequently treated with doxorubicin, vinblastine, bleomycin, and dacarbazine. Thus, the present data allow only conclusions on advanced-stage CHL. Furthermore, the recommendation for using dose-intense treatment strategies is limited to patients with HL younger than age 6 years, because older patients are at increased risk of acute toxicity and treatment-related morbidity when treated with BEACOPP-escalated. 31 Previous reports by our group on elderly patients demonstrated that better tumor control achieved with BEACOPP was offset by higher toxicity and did not translate into better outcome. 32,33 In conclusion, this series of patients with CHL identified differences in disease characteristics, prognostic factors, and treatment outcomes between patients with CHL and other HL entities. In contrast, patients with advanced-stage CHL younger than age 6 years have similar outcomes compared with patients with classical HL when treated with dose-intensified BEACOPP. AUTHORS DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The author(s) indicated no potential conflicts of interest. AUTHOR CONTRIBUTIONS Conception and design: Beate Klimm, Dennis A. Eichenauer, Volker Diehl, Michael Fuchs, Peter Borchmann, Andreas Engert Collection and assembly of data: Beate Klimm, Jeremy Franklin, Harald Stein, Dennis A. Eichenauer, Heinz Haverkamp, Volker Diehl, Michael Fuchs, Peter Borchmann, Andreas Engert Data analysis and interpretation: Beate Klimm, Jeremy Franklin, Harald Stein, Heinz Haverkamp, Andreas Engert Manuscript writing: All authors Final approval of manuscript: All authors REFERENCES 1. Swerdlow SH, Campo E, Harris NL, et al: WHO classification of Tumours of Haematopoietic and Lymphoid Tissues (ed 4). Lyon, France, International Agency for Research on Cancer, Shimabukuro-Vornhagen A, Haverkamp H, Engert A, et al: Lymphocyte-rich classical Hodgkin s lymphoma: Clinical presentation and treatment outcome in 1 patients treated within German Hodgkin s Study Group trials. J Clin Oncol 23: , Nogová L, Reineke T, Brillant C, et al: Lymphocyte-predominant and classical Hodgkin s lymphoma: A comprehensive analysis from the German Hodgkin Study Group. J Clin Oncol 26: , Mani H, Jaffe ES: Hodgkin lymphoma: An update on its biology with new insights into classification. Clin Lymphoma Myeloma 9:26-216, Pileri SA, Ascani S, Leoncini L, et al: Hodgkin s lymphoma: The pathologist s viewpoint. J Clin Pathol 55: , Slack GW, Ferry JA, Hasserjian RP, et al: Lymphocyte depleted Hodgkin lymphoma: An evaluation with immunophenotyping and genetic analysis. Leuk Lymphoma 5: , Hasenclever D: The disappearance of prognostic factors in Hodgkin s disease. Ann Oncol 13: 75-78, 22 (suppl 1) 8. Allemani C, Sant M, De Angelis R, et al: Hodgkin disease survival in Europe and the U.S.: Prognostic significance of morphologic groups. Cancer 17:352-36, Dühmke E, Franklin J, Pfreundschuh M, et al: Low-dose radiation is sufficient for the noninvolved extended-field treatment in favorable early-stage Hodgkin s disease: Long-term results of a randomized trial of radiotherapy alone. J Clin Oncol 19: , Sieber M, Tesch H, Pfistner B, et al: Rapidly alternating COPP/ABV/IMEP is not superior to conventional alternating COPP/ABVD in combination with extended-field radiotherapy in intermediatestage Hodgkin s lymphoma: Final results of the German Hodgkin s Lymphoma Study Group Trial HD5. J Clin Oncol 2: , Sieber M, Tesch H, Pfistner B, et al: Treatment of advanced Hodgkin s disease with COPP/ABV/IMEP versus COPP/ABVD and consolidating radiotherapy: Final results of the German Hodgkin s Lymphoma Study Group HD6 trial. Ann Oncol 15: , Engert A, Franklin J, Eich HT, et al: Two cycles of doxorubicin, bleomycin, vinblastine, and dacarbazine plus extended-field radiotherapy is superior to radiotherapy alone in early favorable Hodgkin s lymphoma: Final results of the GHSG HD7 trial. J Clin Oncol 25: , Engert A, Schiller P, Josting A, et al: Involvedfield radiotherapy is equally effective and less toxic compared with extended-field radiotherapy after four cycles of chemotherapy in patients with earlystage unfavorable Hodgkin s Lymphoma: Results of the HD8 trial of the German Hodgkin s Lymphoma Study Group. J Clin Oncol 21: , Diehl V, Franklin J, Pfreundschuh M, et al: Standard and increased-dose BEACOPP chemotherapy compared with COPP-ABVD for advanced Hodgkin s disease. N Engl J Med 348: , Engert A, Pluetschow A, Eich HT, et al: Reduced treatment intensity in patients with early-stage Hodgkin s lymphoma. N Engl J Med 363:64-652, Eich HT, Diehl V, Görgen H, et al: Intensified chemotherapy and dose-reduced involved-field radiotherapy in patients with early unfavorable Hodgkin s lymphoma: Final analysis of the German Hodgkin Study Group HD11 trial. J Clin Oncol 28: , Borchmann P, Haverkamp H, Diehl V, et al: Eight cycles of escalated-dose BEACOPP compared with four cycles of escalated-dose BEACOPP followed by four cycles of baseline-dose BEACOPP with or without radiotherapy in patients with by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY

7 Lymphocyte-Depleted Classical Hodgkin s Lymphoma advanced-stage Hodgkin s lymphoma: Final analysis of the HD12 trial of the German Hodgkin Study Group (GHSG). J Clin Oncol doi: 1.12/ JCO Borchmann P, Topp MS, Behringer K, et al: Dacarbazine is an essential component of ABVD in the treatment of early favourable Hodgkin Lymphoma: Results of the second interim analysis of the GHSG HD13 trial. Onkologie 33:V448, 21 (suppl 6) 19. Engert A, Borchmann P, Pluetschow A, et al: Dose-escalation with BEACOPP escalated is superior to ABVD in the combined-modality treatment of early unfavorable Hodgkin lymphoma: Final analysis of the German Hodgkin Study Group (GHSG) HD14 Trial. Blood 116:765a, 21 (abstr) 2. Engert A, Kobe C, Markova J, et al: Assessment of residual bulky tumor using FDG-PET in patients with advanced-stage Hodgkin lymphoma after completion of chemotherapy: Final report of the GHSG HD15 Trial. Blood 116:764a, 21 (abstr) 21. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: , Cox DR: Regression models and life tables. J R Stat Soc B 34:187-22, Hasenclever D, Diehl V: Prognostic score for advanced Hodgkin s disease: International Prognostic Factors Project on Advanced Hodgkin s Disease. N Engl J Med 339: , Benharroch D, Levy A, Gopas J, et al: Lymphocyte-depleted classic Hodgkin lymphoma: A neglected entity? Virchows Arch 453: , Miller TP, Byrne GE, Jones SE: Mistaken clinical and pathologic diagnoses of Hodgkin s disease: A Southwest Oncology Group study. Cancer Treat Rep 66: , Kant JA, Hubbard SM, Longo DL, et al: The pathologic and clinical heterogeneity of lymphocytedepleted Hodgkin s disease. J Clin Oncol 4: , Riddell S, Weinerman B, Kemel S, et al: The treatment resistance of lymphocyte depleted Hodgkin s disease. Cancer 46: , Greer JP, Kinney MC, Cousar JB, et al: Lymphocyte-depleted Hodgkin s disease: Clinicopathologic review of 25 patients. Am J Med 81:28-214, Mauch P, Greenberg H, Lewin A, et al: Prognostic factors in patients with subdiaphragmatic Hodgkin s disease. Hematol Oncol 1:25-214, von Wasielewski R, Seth S, Franklin J, et al: Tissue eosinophilia correlates strongly with poor prognosis in nodular sclerosing Hodgkin s disease, allowing for known prognostic factors. Blood 95: , Klimm B, Diehl V, Engert A: Hodgkin s lymphoma in the elderly: A different disease in patients over 6. Oncology (Williston Park) 21:982-99, Ballova V, Rüffer JU, Haverkamp H, et al: A prospectively randomized trial carried out by the German Hodgkin Study Group (GHSG) for elderly patients with advanced Hodgkin s disease comparing BEACOPP baseline and COPP-ABVD (study HD9elderly). Ann Oncol 16: , Halbsguth TV, Nogová L, Mueller H, et al: Phase 2 study of BACOPP (bleomycin, Adriamycin, cyclophosphamide, vincristine, procarbazine, and prednisone) in older patients with Hodgkin lymphoma: A report from the German Hodgkin Study Group (GHSG). Blood 116: , 21 Appendix Table A1. Study Design and Treatment Regimens of GHSG Trials Trial Stage Trial Design HD4 Early favorable EF 4 Gy or EF 3 Gy IF 1 Gy HD5 Early unfavorable 2 COPP/ABVD EF 3 Gy (bulk 1 Gy) or 2 COPP/ABV/IMEP EF 3 Gy (bulk 1 Gy) HD6 Advanced 4 COPP/ABVD IF bulk/residual mass or 4 COPP/ABV/IMEP IF bulk/residual mass HD7 Early favorable EF 3 Gy (36 Gy spleen) 1 Gy IF or 2 ABVD EF 3 Gy (36 Gy spleen) 1 Gy IF HD8 Early unfavorable 2 COPP/ABVD EF 3 Gy (bulk 1 Gy) or 2 COPP/ABVD IF 3 Gy (bulk 1 Gy) HD9 Advanced 4 COPP/ABVD IF bulk/residual mass or 8 BEACOPP baseline IF bulk/residual mass or 8 BEACOPP escalated IF bulk/residual mass HD1 Early favorable 4 ABVD IF 3 Gy or 4 ABVD IF 2 Gy or 2 ABVD IF 3 Gy or 2 ABVD IF 2 Gy HD11 Early unfavorable 4 ABVD IF 3 Gy or 4 ABVD IF 2 Gy or 4 BEACOPP baseline IF 3 Gy or 4 BEACOPP baseline IF 2 Gy HD12 Advanced 8 BEACOPP escalated RT 3 Gy bulk/residual mass or 8 BEACOPP escalated no RT or 4 BEACOPP escalated 4 BEACOPP baseline RT 3 Gy bulk/residual mass or 4 BEACOPP escalated 4 BEACOPP baseline no RT HD13 Early favorable 2 ABVD IF 3 Gy or 2 ABV IF 3 Gy or 2 AVD IF 3 Gy or 2 AV IF 3 Gy HD14 Early unfavorable 4 ABVD IF 3 Gy or 2 BEACOPP escalated 2 ABVD IF 3 Gy HD15 Advanced 8 BEACOPP escalated ( RT 3 Gy to PET-positive residual 2.5 cm) or 6 BEACOPP escalated ( RT 3 Gy to PETpositive residual 2.5 cm) or 8 BEACOPP-14 ( RT 3 Gy to PET-positive residual 2.5 cm) Abbreviations: ABV, doxorubicin, vinblastine, and bleomycin; ABVD, doxorubicin, vinblastine, bleomycin, and dacarbazine; BEACOPP, bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone; COPP, cyclophosphamide, vincristine, procarbazine, and prednisone; EF, extended field; GHSG, German Hodgkin Study Group; HD, Hodgkin s disease; IF, involved field; IMEP, ifosfamide, methotrexate, and etoposide; PET, positron emission tomography; RT, radiotherapy by American Society of Clinical Oncology 7

8 Klimm et al A B Fig A1. Different histologic variants of lymphocyte-depleted classical Hodgkin s lymphoma (CHL): (A) diffuse fibrosis variant of CHL, and (B) reticular variant of CHL. Images courtesy of Donald J. Innes, Jr, MD, Professor of Pathology at the University of Virginia School of Medicine, University of Virginia, Charlottesville, VA by American Society of Clinical Oncology JOURNAL OF CLINICAL ONCOLOGY