A SERMO Oncologist shares insights on the Lymphoma Oral Abstracts presented at ASCO 2015.
“Some of the information presented has practice-changing potential, whereas much of the other data presents promising new approaches in need of more data and time. Some of the data shows the challenges in treating aggressive T cell neoplasms–truly bad players.”
The big abstract of the session was the GADOLIN trial, LBA 8502, presented by L.H. Sehn. The trial compared obinutuzumab-bendamustine (GB) to bendamustine (B) alone for rituximab-resistant relapsed/refractory iNHL. The median age was 63. The end point was PFS. GB was given x 6 cycles, followed by obinutuzumab (G) maintenance for 24 months. B was given for 6 cycles, without maintenance. GB achieved non-reached PFS versus 15 months for B. RR was 63% for B and 69% for GB. OS was similar. Thus, GB is an effective regimen for iNHL patients refractory to rituximab. Toxicity was similar in the arms, grade 3 or higher AEs were 68% for GB and 63% for B. Grade 3 or higher infusion reactions were 9% for GB.
G. Nowakowski discussed the results. He posed the following questions: How long did rituximab resistance last? How would the trial have looked if rituximab maintenance had been used for the patients? Costs are an issue. The control arm would have cost $66k to administer. The GB arm would cost $157k to administer. It does not take into account the cost of earlier subsequent therapy. Thus, ideally we would want to know the whole cost of caring for the patient. It is also unclear if G would have the same benefit if added to another therapeutic regimen (as opposed to B monotherapy).
I also have some comments about the results and design. The cost-effectiveness of this approach is quite murky. It is certainly an expensive regimen. Bendamustine itself is expensive–so the $92k pricetag of the G is even more imposing. The median age in the trial was 63, 10 years younger than the CLL-11 trial which first explored G’s efficacy and toxicity. The toxicity in a more real-world population would likely be higher, including the infusion reactions. There was no maintenance therapy possible in the control arm, so it is hard to tease out how much benefit derived from giving the G maintenance component–if we want to use this regimen, are we all prisoners to using the full, costly G regimen with 2 years of maintenance? The B dose in the control arm was higher than that in the GB arm, so it is possible that some of the benefit of G was obscured by the study. Finally, B alone is not an ideal control arm, which may have helped the GB arm look better. Nevertheless, this trial will likely lead to a practice change with use of G in the second line for iNHL, though some key questions are unanswered.
Abstract 8505, J.S. Abramson presented brentuximab vedotin (BV) + AVD for non-bulky stage I or II Hodgkin lymphoma (HL). The study showed excellent (100% interim CRs) efficacy, though toxicity was increased. This is a radiotherapy-free option for early-stage HL. However, we need the long-term efficacy results before this approach can replace the standard of chemotherapy + XRT for early-stage HL.
Abstract 8506, Nancy Bartlett presented BV + RCHOP for hi-int or high-risk DLBCL (IPI score 3-5). Median age was 67. CD 30+ was defined as 1% or more of cells being CD30+. Peripheral neuropathy was the most frequent AE. ORR was 80%, with median follow-up 6 months. BV-RCHOP appears to be a possible option for high-risk DLBCL, but perhaps combo regimens with BV should omit vincristine or other known neurotoxins. I also go out on a limb and speculate that perhaps DLBCL has some similarities to HL, which is driven by a small percentage of CD30+ cells. Perhaps the BV is helping eliminate some type of Reed-Sternberg-like cell.
Finally, there was the AATT trial, Abs 8507, presented by N. Schmitz. This looked at the challenging problem of peripheral T cell lymphoma (PTCL). It examined prospectively usuing allo versus auto stem cell transplant as consolidation after 1st-line CHOEP. Patients ranged from 18-60 years old and received CHOEP x 4, followed by DHAP as chemomobilization. PBSC harvest was performed on those randomized to the auto arm or those w/o donors randomized to allo. Auto was with BEAM conditioning. Allo used ABLATIVE Fludarabine-Busulfan-Cytoxan conditioning. They had a lot of dropout prior to transplant due to disease progression. They ended up with 17 pts randomized to auto and 20 randomized to allo. Due to donor availability, 23 autos and 13 allos were performed. Only 62% of patients completed the treatment per protocol. EFS curves were not different. OS did not improve with allo, and the trial was stopped early. There were 6 deaths in the allo arm due to treatment-related mortality (TRM). Of the allo patient who got an allo, there was NO relapse mortality. Total non-relapse mortality in the allo arm was 7. Among the autos, relapse mortality was 7 patients. Thus, allo transplant is not yet standard-of-care for PTCL.
Francine Foss discussed the results. Mortality results were consistent with the prior “Nordic” study that used BEAM conditioning and the German study that used Cy-TBI conditioning. Who does poorly with auto? Patients with a high IPI or PIT score, those who do not achieve CR, or those with marrow involvement. TRM has plagued allo arms in studies of txp for PTCL. Retrospective data suggests that non-ablative and ablative approaches are similarly dismal. Perhaps those who do well do so because of lower-risk disease, as opposed to the transplant. However, there is encouraging data for non-ablative (NMA) conditioning from Andrei Shustov at Univ of WA.
Dr. Foss and I discussed the possible role for NMA transplant in PTCL to reduce TRM and retain a graft-versus-lymphoma effect, which was likely to be present in the AATT trial. A NMA or reduced-intensity approach with a slightly higher (e.g. 600 Gy) radiation course might contain chemo-refractory disease. Further investigation of RIC or NMA approaches might yield a regimen with acceptable TRM and retain the GVL effect. In the meantime, reserve allo txp for those with high-risk PTCL and consider auto consolidation for the rest. In the meantime, I hold out hope that PD-1 blockade or CAR T cells may arise on the scene to provide additional options or more efficacious bridges to transplant.
If you are an M.D. or D.O. in the US or UK, please join us on SERMO.