Bleomycin in older early-stage favorable Hodgkin lymphoma patients: analysis of the German Hodgkin Study Group (GHSG) HD10 and HD13 trials

Population-based studies show that about 1 in 4 patients with a first diagnosis of Hodgkin lymphoma (HL) is ≥60 years of age and considered “older.”^1,2  In the absence of large randomized trials in older HL patients, there is no well-established standard chemotherapy regimen and doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) is considered standard of care.^2-6  However, ABVD is associated with severe toxicity in older patients, particularly from bleomycin-induced lung toxicity (BLT).^2,6-9  Therefore, the use of bleomycin has been questioned in older HL patients, especially in early-stage HL.^10,11 

We recently reported results of a large randomized prospective trial comparing ABVD variants omitting bleomycin, dacarbazine, or both in early-stage favorable HL.¹²  Omission of bleomycin resulted in an increased risk of relapse compared with standard ABVD treatment in this trial including patients aged 18 to 75 years. However, whether the benefits of bleomycin in terms of efficacy outweigh the higher risk of severe toxicity in older patients remains unknown.

We therefore investigated the impact of bleomycin in the ABVD regimen in older (≥60 years) early-stage favorable HL patients by comparing feasibility, toxicity, and efficacy of 2×ABVD or 2×AVD followed by 20 or 30 Gy involved-field radiotherapy (IF-RT). In addition, to measure cumulative toxic effects, we analyzed older patients who received 4×ABVD followed by IF-RT in the HD10 trial.¹³ 

The HD10 and HD13 trials conducted by the German Hodgkin Study Group (GHSG) included early-stage favorable HL patients defined as clinical stages I and II without any of the GHSG risk factors.^12,13 

Both trials were approved by the ethics committees of all participating centers and conducted in accordance with the Declaration of Helsinki. Both trials were funded by Deutsche Krebshilfe and the Swiss Federal Government.

Detailed information on both trials has been published previously.^12,13  ABVD consisted of doxorubicin 25 mg/m², bleomycin 10 mg/m², vinblastine 6 mg/m², and dacarbazine 375 mg/m² each administered on days 1 and 15 of a 29-day schedule.

Both trials included pulmonary function testing before enrollment without cutoff-diffusing capacity for carbon monoxide; patients with pulmonary disease prohibiting therapy according to protocol were deemed ineligible. Pulmonary function testing was repeated at follow up or in case of suspected BLT.^12,13  BLT was defined as documented grade 3-4 adverse event of the category “respiratory tract disorders” in relation to bleomycin administration, with severity ranging from dyspnea to interstitial pneumonitis and lung fibrosis.

We compared outcomes and toxicity rates using Fisher’s exact test and analyzed survival end points according to the Kaplan-Meier method. Analyses were done with SAS, version 9.4 (SAS Institute, Cary, NC).

A total of 287 patients (median age, 65 years; range, 60-75) were included and had an equal distribution of patient and disease characteristics (Table 1). A total of 137, 82, and 68 patients were randomized to receive 2×ABVD (HD10 and HD13), 2×AVD (HD13), and 4×ABVD (HD10), respectively.

Early termination of treatment was reported in 2/82 (2%) and 5/137 (4%) of the patients randomized to receive 2×AVD and 2×ABVD, respectively. In contrast, 12/68 (18%) of the patients randomized to receive 4×ABVD terminated therapy early, mainly from toxicity.

Accordingly, patients receiving 2 cycles of chemotherapy were more likely to receive the correct number of chemotherapy cycles and had shorter mean delay of chemotherapy than patients randomized to 4×ABVD (Table 1).

Overall, frequency of grade III-IV adverse events was similar for patients receiving 2×AVD and 2×ABVD (40% and 39%, respectively; Table 1). Importantly, overall grade III-IV toxicity and grade III-IV leukopenia and infection rates were much higher in patients receiving 4×ABVD.

In patients receiving 2 cycles of chemotherapy, respiratory adverse events were rare: BLT was reported in 2 and no cases receiving 2×ABVD and AVD (1.5% and 0%), respectively. In striking contrast, BLT occurred in 7 patients (10%) receiving 4×ABVD and was lethal in 3 of the affected patients (Table 1). This is in line with previous findings reporting BLT rates in older HL patients of about 5% to 36% and an associated lethality of up to 25%.^6,8 

Interestingly, a recent retrospective analysis including older patients with mostly advanced-stage HL found that pulmonary toxicity was the most frequent toxicity with a total incidence of BLT of 27% and that 5 of the 7 patients who died of pulmonary toxicity had received more than 2 cycles of ABVD.¹¹  Similarly, Martin et al reported more BLT in patients treated with ABVD as front-line therapy compared with mustargen, oncovin, procarbazine, and prednisone/ABV(D) combinations and suspected larger exposure to bleomycin in patients treated with ABVD alone as cause.⁹  Furthermore, as recently reported in advanced-stage patients with positron emission tomography–negative restaging after 2 cycles of ABVD, omission of bleomycin resulted in reduced toxicity without loss of efficacy.¹⁴ 

Regarding the efficacy in our study, remission rates and rates of progression and relapse were excellent in all treatment groups. Rates of complete remission ranged between 96% and 99% in the groups receiving 2 cycles of chemotherapy (Table 1) in line with previously reported findings.^2,6,8,11  In contrast, the complete remission rate was only 88% after 4×ABVD, partly resulting from a higher number of deaths during therapy in this treatment group.

Patient numbers in our study were too small for testing on (non-)inferiority of the AVD regimen (Table 1). However, because efficacy of 2×ABVD was not substantially impaired in older patients, we found no evidence for refraining from the efficacy conclusions drawn from the HD13 trial, which included both younger and older patients (Figure 1, Table 1).¹² 

Our findings indicate that the influence of bleomycin on efficacy observed in the HD13 trial should also be considered for the subgroup of older patients and that 2 cycles of ABVD can be safely applied in this presumably more vulnerable patient population. However, our findings do not support the use of bleomycin beyond 2 cycles in ABVD for older HL patients, as would be warranted in certain situations.¹⁵  In older early-stage favorable patients, combined modality treatment should thus be favored over 3 to 4 cycles of chemotherapy alone. In early-stage unfavorable and advanced-stage patients requiring more than 2×ABVD, bleomycin might be omitted in subsequent cycles. In addition, caution is warranted in the use of bleomycin in older patients with risk factors for BLT described previously, such as renal insufficiency, pulmonary radiation, underlying lung disease, tobacco history, and concomitant use of granulocyte colony-stimulating factor.^7,16  However, because of the small number of older patients in our study experiencing BLT, we were not able to confirm these risk factors in our analysis.

As another limitation of our study, comprehensive geriatric assessment, which might help to predict toxicity, was not included in the HD10 and HD13 as it is a mandatory part of ongoing GHSG trials in older HL.

In conclusion, in 219 older early-stage favorable HL patients treated with either 2×ABVD or 2×AVD and both followed by IF-RT, no significant effects of bleomycin on the incidence and severity of adverse events were detectable. In contrast, the additional 68 older patients receiving 4×ABVD had more grade III-IV toxicity and a strikingly higher rate of BLT, suggesting a high risk of severe toxicity and limited benefit in older HL patients receiving more than 2 cycles of bleomycin.

Contribution: B.B. and H.G. designed the research, analyzed and interpreted data, and wrote the manuscript; and K.B., P.J.B., F.H., A.K., R.G., B.v.T., D.A.E., C.B., S.B., M.F., V.D., A.E., and P.B. provided patients, collected and analyzed data, and critically reviewed the manuscript before submission.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Boris Böll, University Hospital Cologne, Department of Internal Medicine 1, Kerpener Strasse 62, 50937 Cologne, Germany; e-mail: boris.boell@uk-koeln.de.