- © 2013 by American Society of Clinical Oncology
ABVD in Older Patients With Early-Stage Hodgkin Lymphoma Treated Within the German Hodgkin Study Group HD10 and HD11 Trials
- Boris Böll,
- Helen Görgen,
- Michael Fuchs,
- Annette Pluetschow,
- Hans Theodor Eich,
- Mario J. Bargetzi,
- Eckhart Weidmann,
- Christian Junghanß,
- Richard Greil,
- Alexander Scherpe,
- Oliver Schmalz,
- Dennis A. Eichenauer,
- Bastian von Tresckow,
- Achim Rothe,
- Volker Diehl,
- Andreas Engert and
- Peter Borchmann⇑
- Boris Böll, Michael Fuchs, Dennis A. Eichenauer, Bastian von Tresckow, Achim Rothe, Andreas Engert, and Peter Borchmann, University Hospital Cologne; Boris Böll, Helen Goergen, Michael Fuchs, Annette Pluetschow, Hans T. Eich, Dennis A. Eichenauer, Bastian von Tresckow, Achim Rothe, Volker Diehl, Andreas Engert, and Peter Borchmann, German Hodgkin Study Group, Cologne; Hans T. Eich, University Hospital Muenster, Muenster; Eckhart Weidmann, Nordwest Hospital Frankfurt, Frankfurt; Christian Junghanß, University Hospital Rostock, Rostock; Alexander Scherpe, Dr. Hancken Clinic, Stade; Oliver Schmalz, Helios Hospital Wuppertal, Wuppertal, Germany; Mario Bargetzi, University Hospital Internal Medicine, Kantonsspital Aarau, Aarau, Switzerland; and Richard Greil, Paracelsus Medical University Salzburg, Salzburg, Austria.
- Corresponding author: Peter Borchmann, MD, Department I of Internal Medicine, University Hospital of Cologne, 50924 Cologne, Germany; e-mail: peter.borchmann{at}uni-koeln.de.
-
Presented in part as oral presentation at the 52nd Annual Meeting of the American Society of Hematology, Orlando, FL, December 6, 2010.
Abstract
Purpose Older patients with Hodgkin lymphoma (HL) account for approximately 20% of all HL patients. ABVD (doxorubicin, bleomycin, vinblastine, and dacarbazine) chemotherapy is regarded as standard of care in these patients. However, little is known on feasibility and efficacy of ABVD in this age group.
Patients and Methods We analyzed the feasibility and efficacy of four cycles of ABVD in older patients age 60 to 75 years with early-stage HL who were treated within the German Hodgkin Study Group (GHSG) HD10 and HD11 trials; results were compared with those of younger patients treated within these trials.
Results In total, 1,299 patients received four cycles of ABVD, and 117 of those patients were older than age 60 years (median, 65 years). In 14% of older patients, treatment was not administered according to protocol, mainly because of excessive toxicity. The mean delay of treatment was twice as high in the older patients (2.2 v 1.2 weeks). Fifty-nine percent of older patients achieved a relative dose-intensity of at least 80% compared with 85% of younger patients. Major toxicity (WHO grade 3 and 4), including leucopenia, nausea, infection, and others, was documented in 68% of older patients with a treatment-related mortality of 5%. Complete response was achieved in 89% of older patients, 3% had progressive disease, and 11% relapsed. At a median observation time of 92 months, 28% of the patients had died, and the 5-year progression-free survival estimate was 75% (95% CI, 66% to 82%).
Conclusion In patients age ≥ 60 years with HL, four cycles of ABVD is associated with substantial dose reduction, treatment delay, toxicity, and treatment-related mortality.
INTRODUCTION
Despite substantial advances in the treatment of Hodgkin lymphoma (HL), 5-year overall survival (OS) in patients age ≥ 60 years (older patients) with HL is as low as 30% to 50%.1,2 At present, approximately 25% of all patients with HL are among the older patients,3,4 and unmet medical needs for these patients will become even more significant in the future as a result of the demographic changes.5 The rather poor prognosis for older patients with HL is due to both patient-specific and lymphoma-associated features. HL in older patients is biologically different and thus more aggressive compared with HL in younger patients.6,7 However, the key factor for the poorer outcome in this group of patients is the drastically increased toxicity of chemo- and radiotherapy resulting in a higher treatment-related mortality and insufficient dosing of the applied treatment.3,8,9
Since most clinical trials exclude older patients with HL, current treatment recommendations are largely based on retrospective population-based studies and small nonrandomized trials. These trials unequivocally show that older patients have a more unfavorable risk profile and require effective treatment to achieve cure.10,11 Approaches to treating older patients with HL with more intensive regimens resulted in treatment-associated mortality of up to 21%.8,9 Therefore, the ABVD (doxorubicin, bleomycin, vinblastine, and dacarbazine) regimen has become widely accepted as standard treatment for older patients with HL, largely on the basis of experiences in younger patients.3,4,12 Toxicity and efficacy of ABVD in older patients with HL has not been evaluated in larger trials thus far. We therefore analyzed the German Hodgkin Study Group (GHSG) HD10 and HD11 trials for feasibility and efficacy of four cycles of ABVD in patients age 60 to 75 years.
PATIENTS AND METHODS
Patients and Study Design
Between May 1998 and January 2003, patients with first diagnosis of favorable or unfavorable early-stage HL according to the GHSG risk stratification were enrolled onto the GHSG HD10 and HD11 trials, respectively.13,14 Both trials were approved by the ethics committees of all participating centers and conducted in accordance with the Declaration of Helsinki.
To investigate feasibility and efficacy of ABVD in older patients, we focused on patients randomly assigned to receive four cycles of ABVD followed by either 30 Gy or 20 Gy of involved field radiotherapy (IFRT; treatment arms A and B in both studies). Results on the remaining patients randomly assigned to receive two cycles of ABVD followed by IFRT (treatment arms C and D in HD10) or four cycles of BEACOPP (bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone) baseline followed by IFRT (treatment arms C and D in HD11) are provided in the Appendix (online only).
ABVD was given on days 1 and 15 in monthly cycles at full dosage if the WBC was at least 2,500/μL and platelets were 80,000/μL on the scheduled treatment day. Treatment was postponed until the indicated levels were achieved, and granulocyte colony-stimulating factor (G-CSF) was given if clinically indicated as judged by the local investigator.13,14 All WHO grade 3 or 4 toxicities were documented. Additional information on both trials including a detailed description of staging and restaging definitions is included in the Appendix.
Study Objectives, Definition of End Points, and Statistical Methods
The focus of this study was the assessment of feasibility, safety, and efficacy of four cycles of ABVD in older patients with HL. Feasibility was measured according to protocol adherence; rates of and reasons for protocol deviation and dose delivery were assessed.
The cutoff values for definition of protocol deviation were more than 25% deviation from the planned total dose, and/or more than 50% deviation from the recommended dose of a single drug, or administration of unknown or considerably more intensive treatment compared with protocol therapy. The relative dose-intensity (RDI) was calculated by dividing the total relative chemotherapy dose by the total relative chemotherapy duration. For the safety evaluation, we analyzed the toxicity of chemotherapy and treatment-associated deaths.
End points for the efficacy analysis were response rates, OS, and progression-free survival (PFS), as described elsewhere.13,14 To adjust for age-related deaths in older patients, we calculated time to HL-related death (OSHL) and time to HL-related failure (PFSHL), defined analogously to OS and PFS, but deaths for known reasons other than HL or toxicity of primary or salvage treatment were censored and not considered as failures.15
Exact CIs were used when appropriate. For comparison of older and younger patients, Fisher's exact test (categorical data) or Wilcoxon rank sum test (continuous data) was used. OS and PFS as well as OSHL and PFSHL were estimated according to the method of Kaplan and Meier, age groups were compared by using log-rank tests, and hazard ratios (HRs) were calculated by using univariate Cox regression. Statistical analyses were performed by using SAS (SAS Institute, Cary, NC), Version 9.2 for Microsoft Windows.
RESULTS
Patient Characteristics
Between May 1998 and January 2003, a total of 1,299 patients, of whom 117 patients were older than age 60 years, were registered and randomly assigned in the GHSG HD10 and HD11 trials. Median age in the older group was 65 years; there were slightly more men, and more than half the patients had mixed cellularity subtype. In the younger reference patients with a median age of 33 years, sex was similarly distributed, and most patients had nodular sclerosis HL. The main patient characteristics are summarized in Table 1.
Patient Characteristics
Feasibility
Eighty-six percent of the older patients in the ABVD group received treatment according to protocol; protocol deviations were documented in 16 patients (14%; 95% CI, 9% to 20%) mainly because of excessive toxicity. In total, 93% of the older patients received four cycles of treatment, 3% received two cycles, and 4% received three cycles.
In HD10, protocol deviation was more frequent in older patients compared with the younger reference group, although rates for older and younger patients in HD11 were comparable. Consequently, the mean total relative chemotherapy dose in older HD11 patients (97%) was comparable to that of the younger population, although it was lower in the elderly HD10 subgroup (93%; Fig 1).
Dose delivery and protocol adherence according to stage and age group. Lines indicate mean relative doses. HD10, favorable early-stage patients from the German Hodgkin Study Group HD10 trial; HD11, unfavorable early-stage patients from the German Hodgkin Study Group HD11 trial.
The mean total treatment delay was 2.2 weeks in older patients; about half the cycles started with a maximum delay of 1 day. Younger patients had a considerably lower mean treatment delay (1.2 weeks). Therapy delays were more frequent than dose reductions in the older HD11 subgroup.
The mean RDI was 83% in the older population; only 59% of patients achieved an RDI of at least 80%. Low RDI was caused by dose reductions and treatment delays. This was comparable in older patients in both studies and considerably less frequent in the younger reference group in which 85% of patients had an RDI of at least 80%. Even within the older subgroup, a strong trend toward lower RDI with higher age was observed: patients age 60 to 64 years, 65 to 69 years, and 70 to 75 years had a mean RDI of 87%, 81%, and 75%, respectively. G-CSF was administered in 25% of the older patients compared with 11% of the younger reference group. The rate was slightly higher in HD11 patients compared with HD10 patients (31% and 21%, respectively). Protocol deviations as well as low RDI were less frequent in patients receiving G-CSF. Low RDI (< 80%) was more frequent in patients with severe leukopenia (71.6% v 40%). Details on dose delivery and protocol adherence are displayed in Table 2. Information on the administration of IFRT is provided in the Appendix.
Feasibility
Acute and Late Toxicity
WHO grade 3 and 4 toxicities during chemotherapy were documented in 68% of the older patients (95% CI, 60% to 75%) and included grade 4 in 18%. Major toxicities were leucopenia, nausea, vomiting, and infection (Table 3). Acute toxicity was less frequent in the first cycle (36%) and varied between 52% and 55% in subsequent cycles. Six patients (5%) died as a result of acute toxicity due to pneumonia with septic multiorgan failure (n = 2), bleomycin-induced lung toxicity (n = 2), cardiac arrhythmia (n = 1), and sepsis (n = 1). There was no difference in toxicity between favorable and unfavorable early-stage patients. In younger patients, toxicity was less common; only 7% experienced grade 4 toxicity, and fewer than 1% died as a result of acute toxicity.
Safety
Within a median observation period of 85 months, secondary malignancies were observed in 15 older patients (13%), including non-Hodgkin lymphoma in four and solid tumors in 11. Solid tumors included colorectal cancer (n = 3), lung cancer (n = 3), breast cancer (n = 2), melanoma (n = 1), prostate cancer (n = 1), and ovarian cancer (n = 1). No secondary leukemia was observed. There was no difference in cancer incidence between favorable and unfavorable early-stage patients; however, all six patients who died as a result of secondary malignancies were in the unfavorable early-stage risk group. In the younger reference group, rates for the incidence of secondary malignancies and death as a result of secondary malignancies were lower (3% and 1%, respectively).
Disease Control and Survival
At final restaging 3 months after the end of therapy, complete response (CR) and unconfirmed CR were achieved in 89% of older patients (95% CI, 83% to 93%). A total of 2% had partial response, and another 2% had progressive disease (PD). All patients with PD had unfavorable early-stage disease. In 8% of the older patients, response could not be assessed because of death after discontinuation of therapy (six patients) or lack of further information (three patients). In younger patients, CR was achieved in 96% and partial response was achieved in 1% (Table 4). Within the HD10 cohort, the CR rate was significantly lower in older compared with younger patients (P = .001), whereas in HD11, there was no significant difference between age groups (P = .2).
Efficacy
With a median observation time of 85 months, 3% of the older patients had PD, and 11% relapsed. Eight of the 15 older patients with PD or relapse had an RDI below 80%. In the younger population, PD (1%) and relapse (8%) were less common. Regarding the single studies, the difference between older and younger patients was more pronounced in favorable early-stage HL, although rates in unfavorable early-stage HL were comparable between age groups.
In total, 33 older patients (28%) died within a median observation time of 92 months; most of these patients had cardiovascular disease (7%), progressive or relapsing HL (5%), or toxicity resulting from chemotherapy (5%). The mortality in unfavorable early-stage patients (37%) was higher compared with that in favorable early-stage patients (22%), mainly because of the high number of deaths attributed to secondary malignancies. Within the same observation period of 92 months, 4% of the younger patients died (Table 3).
The OS and PFS estimates for all older patients at 5 years were 81% (95% CI, 73% to 87%) and 75% (95% CI, 66% to 82%), respectively. The favorable early-stage older subgroup had a better outcome compared with the unfavorable early-stage older subgroup with respect to both end points. In both subgroups, OS and PFS of younger patients were significantly superior compared with those of older patients (all log-rank P < .001; Table 4). The 5-year estimates for OSHL and PFSHL were 88% (95% CI, 80% to 93%) and 82% (95% CI, 74% to 88%), respectively, in all older patients. Regarding those HL-specific end points, the difference between favorable and unfavorable early-stage elderly patients diminished as a result of the non–HL-related deaths not counting as failures. This also resulted in a smaller difference between older and younger patients in HD11 regarding PFSHL (HR, 1.6; 95% CI, 0.8 to 3.1; log-rank P = .2) whereas the estimates for the younger subgroup in HD10 were clearly superior to those for the older subgroup (HR, 4.0; 95% CI, 2.0 to 7.7; log-rank P < .001). Kaplan-Meier plots are shown in Figure 2.
Kaplan-Meier plots of (A) overall survival, (B) progression-free survival, (C) time to HL-related death, and (D) time to HL-related failure according to stage and age group. HD10, favorable early-stage patients from German Hodgkin Study Group HD10 trial; HD11, unfavorable early-stage patients from German Hodgkin Study Group HD11 trial.
DISCUSSION
Knowledge of toxicity and efficacy of ABVD in older patients with HL is astonishingly limited. We therefore analyzed feasibility and efficacy of four cycles of ABVD in 117 older patients with HL who were treated within the GHSG HD10 and HD11 trials. The key findings of this study are as follows: (1) Four cycles of ABVD have considerable toxicity in older patients with HL and resulted in grade 3 to 4 toxicities in more than two thirds of patients (68%). The treatment-related mortality was 5% and might even be underestimated since deaths resulting from cardiovascular events that accounted for 7% mortality might be caused partly by toxicity of chemo- or radiotherapy. (2) Feasibility of ABVD in older patients is poor compared with that in younger patients as indicated by treatment delay, dose reduction, and early termination. Overall, dose delivery was low, with a mean RDI of 83% in the older population; only 59% of older patients achieved an RDI of 80% or higher. (3) Despite high remission rates, progression and relapse occurred in approximately 14% of patients, and survival at 5 years was 10% lower in older compared with younger patients, even after adjusting for mortality not related to HL or treatment.
Regarding the feasibility of ABVD in older HL patients, we found more treatment delay but less dose reduction in unfavorable early-stage (HD11) compared with favorable early-stage (HD10) patients. Since patients in HD10 were randomly assigned between two and four cycles of ABVD, one possible explanation is that treating physicians were less reluctant to reduce treatment in favorable early-stage patients. Accordingly, less early termination due to toxicity occurred in unfavorable early-stage patients, although rates of acute toxicity did not differ between both subgroups. Importantly, low RDI might be one of the key reasons for lower rates of remission and survival in older patients with HL.10 Leukopenia was the most frequent severe toxicity and was associated with a lower RDI. Although the administration of G-CSF with ABVD is controversial, its use according to guidelines might have improved the protocol adherence and RDI in this patient cohort.16–18 But nonhematologic toxicities also occurred more often in older patients. Our observations therefore suggest that the lower RDI in older patients with HL is likely caused by several factors and reflects the general vulnerability of the older patient.
Two important limitations in this analysis include the lack of a comprehensive geriatric assessment and restriction to early-stage patients, which impedes the comparison of our data with previous results on ABVD in older patients with advanced-stage HL.
Trials on ABVD in older patients are scarce; reports have been made on only a few patients, and the information on toxicity and efficacy is heterogeneous. Levis et al19 compared ABVD with VEPEMB (vinblastine, cyclophosphamide, procarbazine, prednisone, etoposide, mitoxantrone, and bleomycin) in 54 patients older than age 65 years with early- and advanced-stage HL; they reported 76% grade 3 or 4 leucopenia as the most frequent toxicity, and contrary to our results, they reported no deaths as a result of toxicity. CR was achieved in 86% of the patients receiving ABVD.
Similar results were recently presented from the United Kingdom-based SHIELD program20 and the United States Intergroup (US Intergroup Trial E2496).21 In these trials, 35 and 23 older patients, respectively, were treated with ABVD. One patient with early-stage HL (total n = 13) experienced fatal bleomycin-induced pulmonary toxicity. In advanced-stage patients, treatment was associated with 18% mortality in the United Kingdom-based trials and and 5% mortality in the US-based trials. Treatment-related mortality was related to septicemia in the SHIELD program. The preliminary US Intergroup Trial E2496 results include grade 3 or 4 bleomycin-associated pulmonary toxicity in 26% of patients and two fatalities.21 This is in accordance with our findings of higher respiratory toxicity in older compared with younger patients (5% v 2%, respectively).
One important limitation of clinical trials in older patients is the exclusion of frail and very old patients. The same applies to the studies reported here, since registry-based data suggest that approximately 20% of all patients with HL are older, but they accounted for only approximately 10% of the population studied. This suggests that only a selected population of older patients who were eligible for evaluation were included in clinical trials and allows only indirect conclusions on the general population of older patients with HL. Nevertheless, one key finding is that even in these carefully selected older patients, efficacy is directly linked to relative dose-intensity, which invariably causes toxicity.3,10 In previous GHSG trials, introduction of more aggressive regimens such as BEACOPP resulted in higher tumor control compared with COPP/ABVD (cyclophosphamide, vincristine, procarbazine, prednisone/ABVD) but caused intolerable toxicity resulting in up to 21% deaths as a result of toxicity in older patients.8,9 Other regimens tested in older patients were also either too toxic or resulted in insufficient lymphoma control.22,23
Kolstad et al24 retrospectively analyzed the use of the well-established CHOP-21 (cyclophosphamide, doxorubicin, vincristine, prednisone, q 21 days) regimen in 29 older patients with HL and reported promising efficacy. However, one death as a result of toxicity was reported; moreover, only limited conclusions on efficacy and toxicity can be drawn because of the small number of patients. More recently, we published the results of a phase II trial that evaluated the PVAG regimen (prednisone, vinblastine, doxorubicin, gemcitabine) in older, mostly advanced-stage (93%) patients with HL, and we observed promising efficacy, which was again accompanied by considerable toxicity.15
Because of the high toxicity with aggressive treatment approaches in older patients, ABVD has become the treatment of choice in these patients on the basis of experience with ABVD in younger patients. However, our results indicate that even four cycles of ABVD have considerable toxicity in older early-stage patients and that efficacy is clearly lower compared with that in younger patients. Both toxicity and lower efficacy of ABVD are likely to be even more pronounced in advanced-stage patients, given their higher tumor burden. In addition, advanced-stage patients are more likely to experience cumulative toxicity, because they receive six to eight cycles instead of four cycles of ABVD. Thus, new strategies with novel drugs are needed for this patient population. Trials combining a truncated ABVD variant (AVD) with lenalidomide (NCT01056679) or brentuximab-vedotin (NCT01476410) in older patients with HL are ongoing and will hopefully result in improved feasibility and efficacy.
In summary, this analysis shows that four cycles only of ABVD are associated with poor feasibility and substantial treatment-related mortality in older patients with early-stage HL. Together with recent observations of ABVD in the treatment of patients with advanced-stage HL, these findings challenge ABVD as standard treatment and underscore the necessity to develop treatment strategies suited for the specific needs of older patients with HL.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Conception and design: Boris Böll, Helen Görgen, Peter Borchmann
Administrative support: Richard Greil
Provision of study materials or patients: Eckhart Weidmann, Richard Greil, Mario J. Bargetzi
Collection and assembly of data: Boris Böll, Helen Goergen, Michael Fuchs, Hans T. Eich, Mario J. Bargetzi, Eckhart Weidmann, Christian Junghanß, Richard Greil, Alexander Scherpe, Oliver Schmalz, Dennis A. Eichenauer, Bastian von Tresckow, Achim Rothe, Volker Diehl, Andreas Engert, Peter Borchmann
Data analysis and interpretation: Boris Böll, Helen Goergen, Annette Pluetschow, Peter Borchmann
Manuscript writing: All authors
Final approval of manuscript: All authors
Appendix
Supplementary Patients and Methods
Trial designs, definition of early-stage Hodgkin lymphoma (HL, and selection criteria for HD10 and HD11.
HD10 included patients with favorable early-stage HL according to the German Hodgkin Study Group (GHSG) definition: clinical stage I and II patients with or without “B” symptoms without any of the following risk factors. HD11 included patients with unfavorable early-stage HL according to the GHSG definition: stage IA, IB, or IIA with at least one of the following risk factors: bulky mediastinal mass (one-third or more of maximum transverse thorax diameter), extranodal involvement, erythrocyte sedimentation rate ≥ 50 mm/h or ≥ 30 mm/h in patients with B symptoms, or three or more lymph node areas involved. In addition, patients with stage IIB disease and erythrocyte sedimentation rate ≥ 30 mm/h or three or more involved lymph node areas but without bulky mediastinal mass or extranodal involvement were included.
For inclusion in both trials, patients were required to have normal organ function as judged by the local investigator and to be in good general condition (WHO index ≤ 2). Patients with impaired heart, lung, liver, or kidney function, previous malignant disease, or positive HIV status were not included. Patients with cardiomyopathy, ischemic heart disease, or heart failure were excluded from both protocols. In addition, patients were excluded from the study if they had chronic obstructive lung disease, were pregnant or lactating, or had HL as part of a composite lymphoma. Minimal hematologic requirements were a WBC count of more than 3,000/μL and a platelet count of more than 100,000/μL. After initial diagnosis by the local pathologist, biopsy material was centrally reviewed by at least one member of a panel of six HL pathology experts.
In the HD10 trial, patients with favorable early-stage disease were randomly assigned to receive either four or two cycles of ABVD (doxorubicin, bleomycin, vinblastine, and dacarbazine), followed by either 20 or 30 Gy involved field radiation therapy (IFRT), in a 2 × 2 factorial design. In the HD11 trial, patients with unfavorable early-stage disease were randomly assigned to receive either four cycles of ABVD or BEACOPP (bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone) baseline, followed by either 20 or 30 Gy IFRT in a 2 × 2 factorial design.
Response Assessment
Final restaging by using computed tomography scans of the involved sites was performed after the end of irradiation, including assessment of the initially involved sites. Complete response (CR) was defined as disappearance of all clinical and radiologic disease; partial response was defined as reduction of at least 50% of maximal diameter compared with the initial involvement. Residual disease after chemo- and radiotherapy was considered as CR unconfirmed with residual lesion if no additional treatment was required within 6 months after the end of protocol treatment.
Supplementary Results
Administration of IFRT.
For the evaluation of IFRT, dropouts before the start of IFRT were excluded, resulting in an analysis set containing 108 of the older (92%) and 1,152 of the younger (98%) patients. Discontinuations of IFRT are documented in three patients in the older subgroup; none of them was caused by acute toxicity. The mean total radiation dose was 24.7 Gy compared with 25.4 Gy in younger patients. Acute toxicities of Common Toxicity Criteria (CTC) grade 3 were observed in five older patients, including dysphagia, mucositis, and local skin reaction. No toxicities of CTC grade 4 were observed.
Results of two cycles of ABVD in older patients with favorable early-stage HL.
Seventy-nine older patients with favorable early-stage HL in the HD10 trial were randomly assigned to receive two cycles of ABVD followed by 20 or 30 Gy IFRT. In these patients, median age was 64 years, and patient characteristics were comparable to those of patients receiving four cycles of ABVD.
Protocol deviations (4%) and reduction of cycles (4%) were less common in patients receiving two cycles ABVD compared with the group receiving four cycles ABVD; the mean relative dose-intensity was 91%. Acute toxicity during chemotherapy was considerably lower with only 38% of patients experiencing WHO grade 2 to 4 toxicities. There were no differences regarding administration and toxicity of IFRT between patients receiving two cycles of ABVD compared with patients receiving four cycles. All but three patients in the group receiving two cycles ABVD (96%) achieved CR as final treatment outcome (two patients had partial response; one patient had progressive disease). Rates of progression or relapse (10%) and death (23%) were comparable in both treatment groups, and the 5-year estimates for overall survival (84%) and progression-free survival (79%) did not differ. Interestingly, no death as a result of cardiovascular disease was observed in older patients receiving two cycles of ABVD. Regarding the HL-specific end points, 5-year estimates were higher in the group with reduced chemotherapy (94% for time to HL-related death and 88% for time to HL-related failure). Consequently, the difference in time to HL-related failure compared with younger patients was remarkably smaller in this treatment group of older patients with favorable early-stage HL treated with two cycles of ABVD and IFRT (hazard ratio [HR], 1.7; 95% CI, 0.8 to 3.6).
Results of BEACOPP Baseline in Older Patients With Unfavorable Early-Stage HL
Fifty-two older patients with unfavorable early-stage HL in the HD11 trial were randomly assigned to receive four cycles of BEACOPP baseline followed by 20 or 30 Gy IFRT. In these patients, median age was 65 years, and patient characteristics were comparable to those of the patients receiving four cycles of ABVD.
Protocol deviations were documented in about one fifth of the patients (21%), mostly because of acute toxicity (17%). However, the mean relative dose-intensity was 87% and thus comparable to that of the ABVD group. Eighty-seven percent of patients experienced WHO grade 3 to 4 acute toxicity during chemotherapy; in 50% of patients, WHO grade 4 was the highest toxicity, and 6% of patients died as a result of toxicity. The rate of CR at final restaging was comparable to that in the ABVD group (87%), and the rates of progressive disease or relapse (10%) and death (29%) were slightly lower. The 5-year estimates for all analyzed end points were slightly higher as compared with the ABVD group.
Footnotes
-
See accompanying editorial on page 1502
-
Written on behalf of the German Hodgkin Study Group, the Swiss Group for Clinical Cancer Research, and the Austrian Arbeitsgemeinschaft Medikamentöse Tumortherapie.
-
Supported by the Deutsche Krebshilfe, the Deutsche Forschungsgemeinschaft, and the Swiss Federal Government.
-
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
-
Trials were registered as NCT00264953 and NCT00265018 at www.clinicaltrials.gov.
