Allogeneic HSCT after PD-1 therapy safe, effective in Hodgkin lymphoma
ORLANDO — Allogeneic hematopoietic stem cell transplantation appeared effective and generally safe for patients with relapsed or refractory classical Hodgkin lymphoma who received prior immune checkpoint inhibitor therapy, according to retrospective study results presented at ASH Annual Meeting and Exposition.
The findings — based on extended follow-up from a large international cohort — should allay concerns that patients treated with PD-1 inhibitors prior to transplant are at increased risk for severe immune-related complications, Reid W. Merryman, MD, attending physician in the lymphoma program at Dana-Farber Cancer Institute, told Healio.
“Allogeneic transplant after anti-PD-1 therapy can be performed safely, but clinicians should be aware that the [posttransplant] course for these patients is different, with higher rates of acute and chronic graft-versus-host disease and potentially lower rates of relapse,” Merryman said in an interview. “In addition, the results indicate the best transplant strategy for these patients includes use of posttransplant cyclophosphamide as part of GVHD prophylaxis.”
Approximately 70% of patients with relapsed or refractory classical Hodgkin lymphoma achieve response to anti-PD-1 blockade, but most patients develop resistance to these agents and relapse within a few years.
Prior studies suggested that allogeneic HSCT after PD-1 inhibitor therapy may be linked to higher rates of immune-related complications, such as veno-occlusive disease, acute GVHD and cytokine release syndrome. These findings prompted an FDA warning that raised a note of caution about using PD-1 blockade as a bridge to allogeneic transplant, Merryman said.
However, small cohorts and short follow-up in those prior studies limited researchers’ efforts to accurately assess long-term outcomes and identify risk factors for early toxicity.
“This is really the only curative treatment for a lot of patients with Hodgkin lymphoma, so we’re left with a very important question: Can allogeneic transplant be performed safely in this population?” Merryman told Healio. “If so, what are the best transplant strategies to reduce toxicity and achieve the best possible outcomes for these patients?”
Merryman and colleagues used databases and medical records to assemble a large retrospective international cohort of patients with classical Hodgkin lymphoma who underwent allogeneic HSCT after PD-1 inhibitor therapy.
Researchers pooled data on 197 patients (median age, 31 years; range, 17-68; 59% male) who underwent transplant between 2014 and 2019.
Patients received a median four (range, 2-11) lines of systemic treatment excluding PD-1 inhibitor therapy. The majority had failed prior autologous HSCT (73%).
Patients received a median 9.5 doses (range, 1-74) of a PD-1 or PD-L1 inhibitor.
Most (88%) had PD-1 monoclonal antibody monotherapy, whereas 11% failed PD-1 monoclonal antibody-based combination therapy and 1% failed PD-L1 monoclonal antibody monotherapy.
Best overall responses to this therapy included complete response (39%), partial response (38%), stable disease (12%) and progressive disease (11%).
Median time from last dose of PD-1/PD-L1 inhibitor therapy to allogeneic transplant was 82 days (range, 17-1,029), and more than half (51%) of patients received intervening systemic therapy.
At the time of transplant, 116 patients (59%) were in complete remission, 61 (31%) were in partial remission, five (3%) had stable disease and 15 (8%) had progressive disease.
The majority of patients received reduced-intensity (55%) or nonmyeloablative conditioning (37%).
Donors included haploidentical (44%), matched unrelated (25%), matched related (24%), mismatched unrelated (5%) and cord blood (2%).
More than half of patients (57%) received posttransplant cyclophosphamide to reduce risk for GVHD.
Researchers estimated OS, PFS, cumulative incidence of relapse, nonrelapse mortality, and incidence of acute and chronic GVHD. They also assessed the association between baseline variables and these outcome measures.
Median follow-up after transplant for survivors was 24.3 months (range, 0.8-72).
At 180 days, cumulative incidence of acute GVHD was 56% (grade 2 to grade 4, 38%; grade 3 to grade 4, 16%).
Three percent of patients developed hyperacute GVHD, defined as onset within 14 days of allogeneic transplant.
Nearly one-quarter (22%) of patients developed noninfectious febrile syndrome/cytokine release syndrome, with most cases being grade 1. Median day to onset was day 3 (range, 0-20).
Three percent of patients developed sinusoidal obstruction syndrome. All cases were severe, with median day of onset being day 13 (range, 2-25).
Univariate analysis revealed two factors predictive of acute GVHD. They were shorter time from last dose of PD-1 therapy to transplant ( 82 days vs. > 82 days, 24% vs. 9%; P = .01) and receipt of intervening salvage therapy (no therapy vs. therapy, 22% vs. 11%; P = .04).
Cumulative incidence of chronic GVHD at 2 years was 38%, and 25% required systemic treatment.
Multivariate analysis revealed several factors associated with risk for chronic GVHD. They included number of non-PD-1 systemic therapies prior to allogeneic HSCT (four or less vs. more than four, 31% vs. 47%; HR = 1.99; 95% CI, 1.16-3.4), donor sex (female vs. male, 47% vs. 33%; HR = 1.77; 95% CI, 1.05-3), receipt of posttransplant cyclophosphamide (yes vs. no, 27% vs. 51%; HR = 0.55; 95% CI, 0.33-0.94), days from last dose of PD-1 therapy to allogeneic transplant ( 82 days vs. < 82 days, 31% vs. 45%; HR = 0.5; 95% CI, 0.29-0.85) and doses of PD-1/PD-L1 monoclonal antibodies ( 10 vs. < 10, 28% vs. 48%; HR = 0.45; 95% CI, 0.26-0.78).
At 2 years, 82% of patients remained alive and 68% remained progression free.
Researchers identified no factors significantly associated with OS.
Multivariate analysis identified disease status at allogeneic HSCT (complete response vs. other, 77% vs. 57; HR = 0.5; 95% CI, 0.29-0.84) and posttransplant cyclophosphamide (yes vs. no, 76% vs. 60%; HR = 0.56; 95% CI, 0.33-0.95) as predictors of PFS.
“There is a lot of heterogeneity in terms of what GVHD prophylaxis regimen people use,” Merryman told Healio. “This study was retrospective and these transplant decisions weren’t made randomly; however, with those caveats, posttransplant cyclophosphamide seemed to be associated with much better outcomes. We found that to be true for patients who had a haploidentical donor and those with other donor types, so it seems to be a strong signal that posttransplant cyclophosphamide is the best GVHD prophylaxis regimen for these patients.”
Researchers reported the following 2-year cumulative incidence rates: nonrelapse mortality, 14%; relapse, 18%; and a composite of GVHD-free survival and FS, 46%.
“It’s always dangerous to compare across studies, but the cumulative incidence of relapse in this population was lower than you would expect and lower than previously reported,” Merryman told Healio. “This suggests perhaps that PD-1 blockade prior to transplant continues to provide some benefit in terms of reducing the rate of relapse.”
Because the study was retrospective and no transplant strategies were decided randomly, so selection bias cannot be excluded, Merryman said.
However, the findings suggest allogeneic transplant can be performed safely after PD-1 monoclonal antibody therapy, and that posttransplant cyclophosphamide is the optimal transplant strategy for patients treated with PD-1 blockade.
One key question remains: How should allogeneic transplant be sequenced for patients with relapsed Hodgkin lymphoma.
“Specifically, should PD-1 blockade be used as a bridge to allogeneic transplant, or should transplant be deferred for patients who are responding to PD-1 blockade,” Merryman said. “With initial reports suggesting excessive early toxicity, there was a move away from allogeneic transplant for these patients. I think our assessment has to be more nuanced.”
On one hand, transplantation after PD-1 blockade is associated with increased risk for severe acute and chronic GVHD, but also likely lower rates of relapse and — in this cohort — high rates of OS and PFS.
“However, delaying transplant appears to reduce risk for GVHD but this assumes relapsed patients can be salvaged and also eliminates the apparent benefit that PD-1 may offer in terms of better disease control after transplantation,” Merryman said. “Given these competing factors, I believe these patients should at least be referred to consider the risk and benefits of transplantation on an individual basis.” – by Mark Leiser
Reference: Merryman RW, et al. Abstract 775. Presented at: ASH Annual Meeting and Exposition; Dec. 7-10, 2019; Orlando.
Disclosures: Merryman reports no relevant financial disclosures. Please see the abstract for all other authors’ relevant financial disclosures.