AUTHORS
Daniel Clancy Trotier, MD, Rina Welkie, MPH2, David Bond, MD, BS3, Kami Maddocks, MD4, Jonathon B. Cohen, MD, MS5, Melissa Kives6, Craig Portell7, Javier Pinilla-Ibarz, MD PhD8, Joanna M Rhodes, MD, MSCE9, Gregory Kirkpatrick, MD, PhD10, Umar Farooq, MD11, Harsh Rohit Shah, DO12, Narendranath Epperla, MD, MS, FACP13, Yazeed Sawalha, MD14
AFFILIATIONS
1 The Ohio State University, Columbus, Ohio, United States
2 The Ohio State University, Columbus, Ohio, United States
3 The Ohio State University, Columbus, Ohio, United States
4 The Ohio State University, Columbus, Ohio, United States
5 Winship Cancer Institute, Emory University, Atlanta, Georgia, United States
6 Winship Cancer Institute, Emory University, Atlanta, Georgia, United States
7 University of Virginia, Charlottesville, Virginia, United States
8 Moffitt Cancer Center, Tampa, Florida, United States
9 Rutgers Cancer Institute, New Brunswick, New Jersey, United States
10 University of Iowa, Iowa City, Iowa, United States
11 University of Iowa, Iowa City, Iowa, United States
12 Huntsman Cancer Institute, University of Utah, Salt Lake City, Utah, United States
13 Huntsman Cancer Institute, University of Utah, Salt Lake City, Utah, United States
14 The Ohio State University, Columbus, Ohio, United States
ABSTRACT
Background: Progression of disease within 24 months of first-line therapy (POD24) is a validated prognostic factor in follicular lymphoma (FL). However, as patients progress to third-line (3L) therapy and beyond, the prognostic impact of POD24 status is unclear. This is important as POD24 is frequently used to stratify risk in clinical trials despite limited validation in this setting. Here, we sought to evaluate the prognostic impact of POD24 in patients with FL initiating 3L treatment.
Methods: Patients with FL who received ≥3 lines of systemic therapy were identified from the Lymphoma Innovations ORIEN Network (LION), a prospective observational study enrolling patients with relapsed FL at 7 US academic centers. POD24 was defined as progression (PD) or starting a new line of treatment within 24 months of initiating first-line systemic therapy. Patients with transformation after 1st (1L) or 2nd (2L) lines of treatment were excluded. Survival outcomes were calculated from the start of 3L therapy. Events for progression-free survival (PFS) were PD or death; events for event-free survival (EFS) were start of new treatment, PD, or death. Univariable and multivariable Cox proportional hazards models (MVA) were constructed to identify prognostic factors for PFS, EFS, and overall survival (OS).
Results: Of 466 patients with relapsed/refractory FL, 186 (40%) patients met inclusion criteria, of whom 70 (38%) had POD24, and 116 (62%) had non-POD24. Baseline characteristics at 3L initiation comparing POD24 vs non-POD24 were median age: 59 vs 61 years, ECOG PS 0–1: 100% vs 94% (missing, n=49), stage III–IV: 80% vs 78% (missing, n=51), >4 nodal groups: 28% vs 28% (missing, n=31), and bulky disease >5 cm: 16% vs 16% (missing, n=30). Patients with POD24 were more likely to have elevated LDH (44% vs 25%, missing, n=76, p=0.04) and shorter median time from diagnosis to 3L therapy (39 vs 103 months, p<.0001). Biopsies were performed prior to 2L in 47% of POD24 vs 63% of non-POD24 (missing, n=14), and prior to 3L in 43% vs 56% (missing, n=12), respectively.
At 1L, treatments for POD24 vs non-POD24 were chemoimmunotherapy (CIT; 67% vs 81%) and single agent anti-CD20 monoclonal antibody (mAb) (31% vs 19%, p=.04). At 2L, treatments included CIT (57% vs 52%), anti-CD20 mAb (19% vs 31%), and lenalidomide-based regimens (7% vs 8%). Overall response rates (ORR) for 2L in POD24 vs non-POD24 were 50% (95%CI: 38-62) vs 60% (95%CI: 51-69, p=.17), and complete response rates (CRR) for 2L were 24% (95%CI: 15-36) and 36% (95%CI: 28-46, p=.09), respectively. Median time from 2L to 3L was 18 months (interquartile range (IQR), 6-31) for POD24 vs 31 months (IQR, 15-58) for non-POD24 (p<.0001).
At 3L, treatments for POD24 vs non-POD24 included CIT in 23 (33%) vs 41 (35%), CAR-T in 16 (23%) vs 14 (12%), CD3xCD20 bispecific antibodies in 4 (6%) vs 7 (6%), lenalidomide-based regimens in 11 (16%) vs 25 (22%), and anti-CD20 mAb in 7 (10%) vs 15 (13%). Response rates to 3L were similar: ORR 69% (95%CI: 56-79) vs 69% (95% CI: 60-77, p=.96) and CRR 50% (95%CI: 38-62) vs 46% (95%CI: 36-55, p=.57). At a median follow-up of 5.1 years, there was no significant difference in 5-year PFS (38% for POD24 vs 45% for non-POD24, p=.42) or EFS (30% for POD24 vs 40% for non-POD24, p=.21). However, 5-year OS was significantly inferior in POD24 patients (70% vs 86%, p=.0396). Of deaths, non-lymphoma-related causes accounted for 48% (10/21) in POD24 vs 33% (5/15) in non-POD24 patients, whereas rates of lymphoma-related deaths were similar: 24% (5/21) vs 26% (4/15) (cause of death missing in 6 of each group, p=0.734).
On MVA adjusting for age, sex, stage, LDH, bulky disease, time from 2L, 1L treatment (CIT vs anti-CD20 mAb), and biopsy prior to 2L/3L, POD24 status was not associated with inferior PFS (HR = 1.0, 95%CI: 0.4–2.7, p=.97) or EFS (HR = 1.7, 95%CI: 0.8–4.0, p=.20) but remained significant for OS (HR = 8.4, 95%CI: 1.3–55.8, p=.0285).
Conclusions: In patients with FL receiving 3L therapy, POD24 status was not associated with response rates, PFS, or EFS. This may reflect the exclusion of patients with prior transformation and increasingly effective 3L treatment options. As such, the utility of POD24-based stratification in the 3L setting and beyond may be limited. Although POD24 was associated with inferior OS on MVA, this survival difference appears to be driven primarily by non-lymphoma-related deaths and may reflect complications related to treatment density over time.
DISCLOSURES
Daniel Clancy Trotier, MD: No relevant conflicts of interest to declare.