Real-World Multicenter Study of PD-1 Blockade in HIV-Associated Hodgkin Lymphoma across the United States

Background: Despite a 5 to 20-fold higher risk of Hodgkin lymphoma (HL) in people with HIV (PWH) than the general population, PWH were excluded from clinical trials that led to the approval of anti-PD-1 agents (PD-1) in HL. PD-1 have overall response rates (ORR) of 65-70% in the general population and are used in all lines of HL therapy. Several prospective and retrospective studies have shown PD-1 are safe in PWH, but they remain underutilized in HIV-associated HL (HIV-HL) where data are limited.

Methods: Patients (pts) with HIV-HL were identified from the retrospective, international database, “Cancer Therapy using Checkpoint inhibitors in PWH-International (CATCH-IT)” consortium, that is maintained at the Dana-Farber Cancer Institute (DFCI) and approved by DFCI and local institutional review boards. Pts who received either nivolumab or pembrolizumab, alone or in combination with other agents were included. Records were reviewed for demographics, HL and HIV characteristics, and treatment data.

Safety was evaluated by the occurrence of immune-related adverse events (irAEs) per Common Terminology Criteria for Adverse Events (CTCAE) v5.0. Response to treatment was evaluated using descriptive statistics, and immune reconstitution was measured by changes in CD4⁺ T cell counts (CD4) from prior to end of PD-1 therapy via paired t-tests. ORR was defined as the proportion of pts with partial or complete response (PR/CR) per the 2014 Lugano classification for HL. Duration of response (DOR) was defined as the time from the date of the first CR or PR to the date of progressive disease (PD) or death. Progression-free survival (PFS) was defined as the time from PD-1 initiation to disease progression, death, or censored on the date of the last follow-up. Overall survival (OS) was defined as the time from PD-1 initiation to death or censored at the last follow-up. DOR, PFS, and OS were evaluated by Kaplan-Meier methodology.

Results: We identified 23 pts with HIV-HL from 12 institutions in the United States: 20 cisgender men, 3 cisgender women; 10 Black,10 White, and 10 Hispanic; median age 48 years [interquartile range (IQR): 36-58]. At HL diagnosis, 19 (83%) pts had advanced stage disease, and in 20 pts where EBV status of the tumor was known, all were positive. Nine (39%) pts had primary refractory disease. The median number of prior systemic therapies was 2 (IQR: 1-3). One pt had PD-1 as 1^st-line therapy, 6 had PD-1 as 2^nd-line therapy, and 16 had PD-1 as ≥3^rd-line therapy. Fourteen (61%) pts received doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) as 1^st-line therapy. Two pts had prior autologous stem cell transplant (ASCT). At the time of PD-1 initiation, 22 pts (96%) had advanced stage disease, 17 (46%) had extranodal disease, and the median ECOG performance status was 1 (IQR: 0-2). Seventeen (74%) pts received PD-1 monotherapy, 5 (22%) received nivolumab + brentuximab vedotin, and 1 received nivolumab + ifosfamide, carboplatin, and etoposide. Four pts had uncontrolled HIV at the time of PD-1 initiation (HIV viral load range=10,000-2,000,000 copies/mL). All pts received antiretroviral therapy during treatment. Median baseline CD4 was 155 cells/µL (IQR: 66-297) and CD8 was 429 cells/µL (IQR: 158-716). Median CD4 and CD8 increased to 310 and 653 cells/µL at the end of PD-1 treatment (P=0.0084, P=0.32).

The median number of PD-1 cycles was 6 (IQR: 4-21). Three (13%) pts had irAEs (all grade 3): hypothyroidism, autoimmune pancreatitis, and pneumonitis. None required PD-1 discontinuation. The ORR was 83% in the overall cohort. Sixteen pts (70%) had CR, 3 (13%) PR, and 4 (17%) PD as best response. Among the 19 pts with CR or PR, the median DOR was 19.7 months; 8 had PD after initial response and 11 had ongoing CR at data cut-off. Ten pts received consolidation with ASCT after PD-1, 6 of whom had received PD-1 at ≥3^rd line treatment. All pts who received ASCT had ongoing CR at data cut-off. Median PFS was 21.2 months in the overall cohort. Among pts with CD4 <200 and >200 cells/µL, PFS was not statistically different (17.3 vs 29 months; P=0.95). At data cut-off, 17 pts (74%) were alive and 6 (26%) had died due to HIV-HL. Median OS was 23.6 months in the overall cohort.

Conclusion: In HIV-HL, PD-1 blockade was safe, had a high ORR and CR rate, and allowed for immune reconstitution in pts across a wide range of CD4 counts and HIV viral loads. These data add to the evidence that anti-PD-1 agents are effective in PWH and should be used in HIV-HL.