Background: Response-adapted therapy aims to minimize toxicity while maintaining efficacy. Patients with HIV-associated lymphomas are at increased risk for complications and minimizing treatment is advantageous. In 33 patients with HIV-associated diffuse large B-cell lymphoma (DLBCL), we previously reported a 5-year freedom from progression (FFP) and overall survival (OS) rate of 84% and 68%, respectively, with 79% of patients receiving only 3 cycles of short course EPOCH-RR (Dunleavy, 2010). In that study, treatment duration was based on an interim FDG PET-CT scan comparing global reduction of SUV to baseline. Herein, we report an updated analysis including an additional 22 patients and investigate the predictive value of interim PET scans using modern Lugano response criteria.
Methods: Eligible patients had untreated HIV-associated DLBCL with adequate organ function. All performance status and stages were eligible, including CNS involvement. Pre-treatment evaluation included bone marrow biopsy, lumbar puncture, CT and PET scan, and brain MRI/CT if indicated. Infusional EPOCH was delivered every 21 days with rituximab on days 1 and 5 (SC-EPOCH-RR) without dose-adjustment for a minimum of 3 and maximum of 6 cycles. HIV antiretroviral therapy (ART) was held until completion of systemic chemotherapy. Treatment response was determined by CT and PET imaging after 2 cycles, with restaging imaging performed consecutively after each cycle. Patients that had a global PET decrease > 50% compared to pre-treatment PET, were treated with only 1 additional cycle. Patients who did not decrease > 50% on PET, continued treatment until there was < 25% reduction in bidimensional products on serial CT imaging. Active CNS disease was treated with twice weekly intrathecal (IT) methotrexate until CSF clearance followed by maintenance IT therapy. All other patients received prophylactic IT methotrexate starting on cycle 3 for 8 doses. The primary endpoint of the study was progression-free survival (PFS). Independent and blinded review of PET scans using Lugano response criteria (Deauville score 1-3 = negative, score 4-5 = positive) was performed by a nuclear medicine specialist (M.A.).
Results: Fifty-five patients were enrolled between March 2001 and February 2019. Median age was 42 (range, 9-60) and 85% were male. Ann Arbor stage was III/IV in 84% and 71% had a high-intermediate or high IPI score. Six (11%) patients had CNS involvement, and 6 (11%) had bone marrow involvement. Median CD4 count was 210 cell/m3 (range 0-1192). Cell-of-origin by Hans classified 69% as GCB and 22% were EBV positive. 31 patients were tested for MYC rearrangement by FISH and 5 (16%) were positive. Patients received a median of 3 cycles of therapy; 1 (2%) patient received only 1 cycle, 2 patients (4%) received 2 cycles, 45 patients (82%) received 3 cycles, 3 (5%) patients received 4 cycles, and 4 (7%) patients received 5 cycles. After a median potential follow-up of 12.6 years, 5-year FFP, PFS and OS were 76.0%, 64.3% and 71.8%, respectively [Figures 1, 2]. Two patients progressed during therapy, while 11 patients relapsed after therapy - all within 12 months of treatment completion. Of 18 deaths on study, 6 were due to disease progression and 12 died from causes unrelated to lymphoma or treatment. Of forty-seven patients with interim (post-cycle 2) PET scans, 30 (64%) became PET negative after 2 cycles. The 5-year PFS was not different in patients who achieved interim PET negativity: 73.4% versus 62.7% (p = 0.14 for comparison of entire PFS curves) [Figure 4]. Of the 13 total patients who progressed or relapsed, 5 were negative on interim PET.
Conclusion:Response-adapted SC-EPOCH-RR therapy is effective for HIV-associated DLBCL and many patients are cured with only 3 cycles of therapy. Interim PET scans after 2 cycles using Lugano criteria did not reliably predict clinical outcomes. Further studies investigating the role of circulating tumor DNA (ctDNA) for genotyping and predicting clinical outcomes are planned.
This work was supported by the Intramural Research Program of NCI.
Dunleavy:Pharmacyclics: Membership on an entity's Board of Directors or advisory committees.
Author notes
Asterisk with author names denotes non-ASH members.