Brentuximab vedotin and nivolumab for untreated patients with Hodgkin lymphoma: long-term results Open Access

TO THE EDITOR:

For half a century, ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) has remained the most commonly used regimen for advanced Hodgkin lymphoma with cure rates approaching 90%; however, outcomes for the 15% of patients aged ≥60 years is significantly inferior and associated with greater toxicity, notably from the bleomycin and doxorubicin. The activity with the antibody-drug conjugate brentuximab vedotin (BV)¹ and the check point inhibitors (CPIs) nivolumab² (nivo) and pembrolizumab³ in the relapsed and refractory setting, stimulated their study in the front line in elderly patients.

We previously published a multicenter study with BV-nivo administered to 46 previously untreated patients either 60 years of age or older, or not considered suitable for adriamycin-containing therapy.⁴ Despite the overall response rate (ORR) of 61% and complete response rate (CR) of 48%, the results did not meet the predetermined activity criteria of 80% ORR and was terminated prematurely. At a median follow-up of 20.2 months, the median progression-free survival (PFS) was 18.3 months with 91% still alive at 12 months follow-up. This report describes the long-term follow-up of these patients at a median follow-up 58.5 months.

The methods have been previously published.⁴ In brief, patients were considered eligible if they were either 60 years of age or older and/or were considered unsuitable candidates for standard chemotherapy because of cardiac, pulmonary, or renal dysfunction as detailed in the original publication.⁴ Informed consent was obtained for all patients prior to therapy, with institutional review board approval at all participating institutions. Patients received BV at 1.8 mg/kg (dose cap at 180 mg) and nivo at 3 mg/kg, both IV, every 21 days for 8 cycles. The primary end point was ORR.

Of the original 46 patients, updated follow-up data were available for the 39 patients who were still alive. Median age was 71.5 (range, 46-87) years. There were no differences between those who did or did not require subsequent treatment. The reasons participants were considered ineligible for standard chemotherapy was age ≥60 years in 95.7%, equally distributed between the arms, with 1 patient having cardiac disease, and a single patient with pulmonary disease.

The regimen was reasonably well-tolerated. Reasons for discontinuation of BV-nivo were completion of protocol therapy (70.8%), adverse event (12.5%), patient refusal (8.3%), disease progression (4.2%), and death on study (4.2%).

With a median follow-up now of 58.5 months (range, 0.45-65.4) months, 27 (58.7%) patients had progressed with the median PFS of 18.3 (range, 12.8-46.3) months (Figure 1), whereas 19 (41.3%) had still not experienced a progression event. Median duration of response was 17.2 (range, 11.6-45.2) months, and median overall survival (OS) was not yet reached with 39 (84.8%) patients who were still alive (Figure 2).

Subsequent treatments received by patients not cured with BV-nivo included ABVD (6), AVD (6), local radiation (3), cyclophosphamide, doxorubicin, vincristine, prednisone (CHOP) (3), chemotherapy not otherwise specified (2), and 1 each of pembrolizumab, nivo, BV-nivo, GVD (gemcitabine, vinorelbine, doxil)-nivo, GND (gemcitabine, navelbine, doxorubicin), GDP (gemcitabine, dexamethasone, cis-platin), doxorubicin, brentuximab vedotin, vinblastine, dacarbazine (AAVD), CEOP (cyclophosphamide, etoposide, vincristine, prednisone) + radiation, bendamustine-BV, clinical trial.

Although treatment of advanced Hodgkin lymphoma has been rapidly evolving, outcomes for older patients with standard regimens are generally inferior to younger patients.⁵ Nonchemo agents such as BV and CPI have stimulated interest in their use as initial treatment in this unfavorable group of patients. Several investigators have attempted to incorporate BV or CPI, or to minimize or even eliminate chemotherapy in this population. In a multicenter phase 2 trial, Evens et al sequenced BV with AVD.⁶ Patients received BV × 2, AVD × 6, then BV × 4 in responding patients. The CR rate was 95% with a 2-year PFS of 84%. Even in patients unable to complete therapy due to intolerance, 75% remained free of disease. Registry analyses suggest that full dose chemotherapy is effective in older patients who are considered able to tolerate it, unlike in the current study.^7,8 

The ECHELON-1 regimen established BV-AVD as a new standard in frontline patients with advanced disease.⁹ Recently, the results from SWOG-1826 comparing BV-AVD with nivo-AVD, showed improved PFS and fewer survival events with the nivo-AVD.¹⁰ In ECHELON-1 where BV-AVD improved PFS and OS overall, there was no benefit in patients ≥60 years.¹¹ In contrast, in SWOG S1826, a benefit was achieved in the older patients.¹² In addition, there was less toxicity in the nivo arm, notably significantly fewer episodes of sepsis and infections with nivo, and numerically more neurotoxicity with BV.

Neither nivo nor BV as single agents have shown promising activity in the front line.¹³ 

Forero-Torres et al¹⁴ and Friedberg et al¹⁵ treated 27 patients aged ≥60 years with BV alone. The ORR was 92% with 73% CR; however, the median duration of response was a mere 9.1 months, with a PFS of 10.5 months, and the median OS was not yet reached.

Friedberg et al¹⁶ recently updated their prior study of combinations of BV with dacarbazine (DTIC) or nivo. In 22 patients, BV-DTIC achieved an ORR of 95% with 64% CR; at a median follow-up of 63.6 months, the median duration of response (DOR) was 46 months, median PFS was 47.2 months, and the OS was not yet reached. Of 21 patients who had received BV-nivo, the ORR was 86% with 67% CR. At a median follow-up of 51.2 months, the median PFS, DOR, and OS had not yet been reached. The disparity between these data and those of Cheson et al⁴ was unclear as the patients were of a similar age and stage distribution. However, more patients in the current study had performance status (PS) of 2 and in the present study patients received eight 21-day cycles compared to Friedberg et al in which BV-nivo was administered for up to 16 cycles.

The present study, to our knowledge, was the first and largest series combining BV-nivo in untreated patients and has the longest follow-up of any published series. Limitations of the present results primarily relate to the lack of a standardized comorbidity score, and duration of therapy. In addition, subsequent therapy was not regulated. Nonetheless, despite the suboptimal ORR, the OS was 84.8% at almost 5 years, certainly superior to what would be expected in a similar population.

Despite the disappointing 61% ORR in our study, 42% of the patients still had not progressed out to 5 years and were likely cured with this nonchemo approach. Not only did BV-nivo not compromise the response to subsequent therapy, it may have enhanced it with >80% of patients still alive. The nonchemo program may have improved the condition of patients initially considered poor candidates for chemotherapy, such that those who required subsequent treatment were better able to tolerate and respond to it. Second, the CPI may have sensitized patients to subsequent chemotherapy.^17,18 The BV-nivo combination has now been demonstrated in 2 studies, to achieve durable responses in a significant proportion of older patients. Thus, a possible approach in this patient population would be to treat with BV-nivo to potentially cure a significant number of patients with a nonchemo approach. Those who required subsequent treatment could then do so with the expectation of a successful outcome.

Despite impressive results with novel agents combined with chemotherapy, there is still considerable toxicity. With optimal dosing and scheduling, and the availability of newer therapies, the goal of an effective chemotherapy-free approach should be achievable in this patient population. This trial was registered at www.ClinicalTrials.gov as # NCT02758717.

Contribution: B.D.C., S.M.A., and B.K. conceived the study and was responsible for the statistical analyses; and B.D.C., N.L.B., H.J.L., R.H.A., B.C., C.S.D., T.A.F., and S.M.A. contributed to patient enrollment, provided input and gave final approval for the manuscript.

Conflict-of-interest disclosure: B.D.C. serves on the advisory boards of AstraZeneca and Regeneron; receives research support (to institution) from AbbVie, Genmab, Genentech, MorphoSys, Prelude, Lilly, and BeiGene; and receives speakers bureau from Lilly. N.L.B. serves on the advisory boards of and receives research support (to the institution) from Pfizer/Seattle Genetics. B.K. is a Hodgkin lymphoma consultant for Century Therapeutics; receives honorarium from Aptitude Health, Cancer Experts, Curio Sciences, and Janssen; and receives research support (to the institution) from Bristol Myers Squibb, Celgene, Octernal, Pharmacyclics, Seattle Genetics, and Takeda. R.H.A. serves on the advisory boards of Roche and Merck and receives research funding (to the institution) from Seattle Genetics, Merck, BeiGene, and Roche. B.C. serves on the advisory boards of Ipsen and AstraZeneca and receives research funding (to the institution) from Genentech, Acerta, Millenium, Bristol Myers Squibb, and Kite. C.S.D. provides consultancy for Bristol Myers Squibb and Merck. T.A.F. provides consultancy for ADC Therapeutics, AstraZeneca, Bristol Myers Squibb, Epizyme, Inc., Genmab, Seattle Genetics, Pharmacyclics, and Gilead; serves as a speaker for Seattle Genetics, Genmab, AbbVie, AstraZeneca, Bristol Myers Squibb, Pharmacyclics, and Gilead; and has equity in Genomic Testing Cooperative and OMI. S.M.A. receives research support (to the institution) from Bristol Myers Squibb, Takeda, Pfizer, AstraZeneca, Regeneron, ADC Therapeutics, and Affimed.

Correspondence: Bruce D. Cheson, Center for Cancer and Blood Disorders, 11028 Waycroft Way, Bethesda, MD 20852; email: bdcheson@gmail.com.