Recently, a number of novel therapies have been under investigation in PTCLs, however, it remains a challenge to compare these agents and determine the impact on outcome. The purpose of this population-based study was to determine the spectrum of survival in PTCL patients (pts) following relapse and to explore factors influencing survival. The three most common subtypes encountered in North America were evaluated: PTCL-not otherwise specified (PTCL-NOS), angioimmunoblastic T-cell lymphoma (AILT), anaplastic large cell lymphoma, ALK positive (ALK-pos) and ALK-negative (ALK-neg).
The Centre for Lymphoid Cancer Database was screened to identify all pts ≥ 16 y with the above histologies that relapsed or had progressive disease (PD) following primary therapy. In addition to the timing of relapse (very early < 9 mos from diagnosis; early 9–24 mos and late > 24 mos), clinical information at the time of relapse was collected where possible. Responses were determined by the treating physician: Response = any clinical and/or radiographic response; PD = no change or disease progression.
In total, 276 pts diagnosed between 1976 and 2010 were identified with primary progressive or relapsed PTCL with initial diagnoses by the WHO classification. 68 were excluded: unconfirmed pathology (6), CNS disease (5), complications during primary CHT (17), no primary chemotherapy (CHT) (27), incomplete chart or lost follow-up (13). 208 pts were analyzed (PTCL-NOS = 109 (52%), ALCL = 54 (26%) (ALK-pos = 18; ALK-neg = 33; ALK status unknown = 3), AILT = 45 (22%). The majority of pts received anthracyclines as part of their primary CHT (89%). The median age at the time of relapse was 62.5 y (range 20 to 86 y) (PTCL-NOS 62 y, ALK-neg 62 y, ALK-pos 40.5 y, AILT 68 y). 129 (62%) pts were refractory to initial CHT. 53 pts were planned for HDC/SCT, however, only 38 (72%) ultimately received it (allo 17, auto 21). 37 pts (18%) were either too ill for any therapy (34) or refused therapy (3). The remaining pts received some type of treatment for their relapsed PTCL: systemic CHT (gemcitabine, CHOP(like), alkylators alone) (103) radiation alone (19) or steroids (11).
The median follow-up for surviving pts after relapse/progression was ∼ 4 y. The median overall survival (OS) following relapse for the whole cohort was 7.0 mos (HDC/SCT 47.1 mos; no HDC/SCT 5.4 mos). The median progression free survival (PFS) following relapse was 4.6 mos (HDC/SCT 27.7 mos, no HDC/SCT 2.8 mos). The 3 y OS and PFS following relapse were 55% and 48%, respectively in the HDC/SCT group and similar whether an autologous or allogeneic SCT was performed. The corresponding 3 y OS and PFS estimates in the no HDC/SCT group were19% and 13.5 %, respectively. There was no difference in OS and PFS amongst the histologic subtypes. Considering the no HDC/SCT pts who received CHT (n=103), 55% were determined to have to have had a response to chemotherapy, and the median OS and PFS was 7.0 mos and 4.0 mos, respectively (responders vs non-responders/PD OS 16.7 mos vs 3.0 mos, p<.0001; responders vs non-responders/PD PFS 8.2 mos vs 1.4 mos, p<.0001).
In multivariate analysis, the secondary IPI (sIPI), PS ≥ 2, late relapse and HDC/SCT were all prognostic for both PFS (sIPI ≥3 (HR 1.8, p=.009), PS ≥2 (HR 2, p =.002), HDC/SCT(HR.28, p<.0001), late relapse > 24 mos (HR=.64, p<.0001)) and OS (sIPI (HR 1.84, p=.010), PS (HR 2.5, p<.0001) HDC/SCT (HR.46, p=.003) late relapse (HR.65, p<.0001)). Of interest, 28 pts relapsed > 24 mos from the date of diagnosis (PTCL-NOS 12 (43%); ALK-neg ALCL 6 (21%); ALK-pos ALCL 3 (11%); AILT 6 (21%)) and only 8 (31%) subsequently received HDC/SCT but the median PFS and OS was very favourable in this group, 25.4 mos and 41.4 mos, respectively (no HDC/SCT PFS 14 mos and OS 32.3 mos).
The outcome of the majority of pts with relapsed/progressive PTCL is poor and with brief remissions to systemic therapy, highlighting the urgent need for novel agents specifically active in PTCL. Outcomes were far superior in pts who were able to be transplanted and efforts need to focus on expanding the number of pts eligible for this curative therapy. Standard clinical factors were prognostic at relapse and are critical when comparing the efficacy of new agents. PTCL pts with rare, very late relapses have a much more favourable prognosis and further highlight the diverse disease biology of PTCLs.
Sehn:Roche/Genentech: Consultancy, Honoraria, Research Funding. Villa:Roche: Research Funding. Shenkier:Roche: Research Funding. Savage:Celgene: Consultancy; Allos Therapeutics: Consultancy, Honoraria; Seattle Genetics: Consultancy, Honoraria.
This icon denotes a clinically relevant abstract
Asterisk with author names denotes non-ASH members.