Treatment Patterns and Outcomes for Patients with Classic Hodgkin Lymphoma (cHL) and Cardiomyopathy with Low Ejection Fraction (EF): Real-World Evidence (RWE) from 16 US Academic Centers

Introduction
The cornerstone for management of newly diagnosed patients (pts) with cHL is an anthracycline (AC) based chemotherapy (chemo) regimen. There are a paucity of data available regarding treatment patterns and outcomes for cHL pts with low EF. Furthermore, cHL pts with low EF are typically excluded from clinical trials limiting our understanding on management of these pts.

Methods
This was a multicenter retrospective cohort study evaluating pts with newly diagnosed cHL with low EF from 16 US medical centers. Eligible pts included adults diagnosed after 1/1/2010 with pretreatment EF of <50% on echocardiogram. Pts who developed reduced EF during or after first-line treatment were excluded. Based on EF, pts were divided into 2 groups (gp): <45% (gp 1) and 45-49% (gp 2). First-line treatment was categorized into AC based and non-AC based therapies. Primary outcome was progression-free survival (PFS). Secondary outcomes included response rates to first-line therapy and overall survival (OS). Time to event endpoints were evaluated by Kaplan-Meier and Cox proportional hazard methods.

Results
The study included 69 pts with a median age of 55 yrs (21-88), 74% males, and performance status (PS) of ECOG 0-1 in 65%. There were 37 (54%) pts in gp 1 and 32 (46%) in gp 2. 39 pts received AC and 30 received non-AC based therapies. Pts in the AC group were significantly younger (median age, 46 vs 64 yrs), had higher proportion with EF of 45-49% (72% vs 13%) & ECOG PS of 0-1 (82% vs 54%) compared to non-AC based therapies.

Among the 30 pts who received non-AC based therapies, 26 were in gp 1 and 4 in gp 2. Non-AC based therapies included BV monotherapy (n=6), non-AC BV-based therapies (n=11), and other non-AC chemo (n=13). The most common BV based therapy was BV+dacarbazine (n=4) followed by BV+bendamustine (n=3), while the most common other non-AC chemo was MOPP (n=4) followed by ChlVPP (n=3). The overall response rate (ORR) and complete response rate (CRR) to AC based therapy was 77% (30/39) and 69% (27/39), respectively. Among the non-AC therapies, the ORR/CR rates for BV monotherapy, BV based therapies, and other non-AC chemo were 33%/17%, 55%/27%, and 61%/54%, respectively.

The median PFS for the entire cohort was 2.7 yrs (95%CI= 0.80-5.01). The median PFS for gp 1 vs gp 2 was 1.55 yrs and 4.20 yrs, respectively (p=0.26). The median PFS was significantly longer in pts receiving AC based compared to non-AC therapies (5.01 yrs vs 0.55 yrs, <0.001, Figure 1) with 3- and 5-yr PFS estimates of 62% vs 27%, and 54% vs 16%, respectively. Among the recipients of non-AC therapies, the 3-year PFS estimates for BV monotherapy, BV based therapies, and other non-AC chemo was 20%, 34%, and 42%, respectively (p=0.25). The median OS for the entire cohort was 5.6 yrs (95% CI=3.47-NR). The median OS for gp 1 vs gp 2 was 5.41 yrs and 5.73 yrs, respectively (p=0.82). The median OS was significantly longer in pts receiving AC based compared to non-AC therapies (7.43 yrs vs 3.20 yrs, p=0.003) with 3- and 5-yr OS estimates of 80% vs 60% and 80% vs 30%, respectively. Among the recipients of non-AC therapies, the 3-year OS estimates for BV monotherapy, BV based therapies and other non-AC chemo were 44%, 78% and 51%, respectively.

Univariable Cox models tested baseline pt and disease variables and found increasing age, ECOG PS ≥2, and the presence of geriatric syndrome to be prognostic of inferior PFS and OS but not EF gp. Only increasing age remained associated with significantly inferior PFS and OS in multivariable analysis (Table 1). After adjusting for these variables together with the inclusion of first-line treatment (AC vs non-AC based therapies), we found first-line treatment with AC-based therapy was associated with superior PFS (aHR= 0.39, 95% CI=0.15-0.97, p=0.04) but not OS (aHR= 0.51 (95% CI=0.16-1.65, p=0.26, Table 1). With inclusion of treatment, only age remained associated with inferior OS in the multivariable adjustment (aHR=1.27, 95%CI=1.04-1.54).

Conclusion
This first RWE of cHL pts with low EF found that a majority of pts with EF of 45-49% were treated with AC-based therapies. Furthermore, PFS was improved for pts who received AC-based therapies. However, outcomes appeared modest for pts compared with expectant outcomes. Collectively, these data underscore the critical need for prospective interventional studies in this challenging pt population. Further data on AC dosing, safety, risk of AC-induced heart failure, etc. will be presented at the meeting.