B-Symptoms, Performance Status and Response to First Therapy As Survival Predictive Factors for Peripheral T-Cell Lymphoma at Progression/Relapse: A Single Institute Analysis

Introduction

Peripheral T-cell lymphoma (PTCL) is a heterogeneous group of generally aggressive lymphomas that account for 5-15%of non-Hodgkin lymphomas. PTCL, not otherwise specified (PTCL-NOS), is the most common subtype that accounts for 25.9% of PTCL, followed by angioimmunoblastic T-cell lymphoma (AITL), NK/T-cell lymphoma (NKTCL), Adult T-cell Leukemia/Lymphoma (ATLL), ALK-positive anaplastic large cell lymphoma (ALK+ ALCL), and ALK-negative ALCL (ALK- ALCL) (Vose J. et al, 2008). The Prognostic Index for PTCL-NOS (PIT; Gallamini A. et al, 2004) is a prognostic model widely used for PTCL-NOS. There are, however, no established risk factors developed specifically for PTCL after progression/relapse. We report the survival outcomes of 6 common subtypes of PTCL after first line therapy as well as after first progression/relapse, and report newly identified prognostic factors for PTCL after progression/relapse.

Methods

We identified 144 patients with newly diagnosed PTCL-NOS, AITL, ALK+ ALCL, ALK- ALCL, NKTCL and ATLL between 2005-2017 in our institution. After excluding patients who did not receive treatment in our institution, we performed a retrospective analysis of the remaining 132 newly diagnosed patients (PTCL-NOS n=47, AITL n=21, ALK- ALCL n=6, ALK+ ALCL n=9, NKTCL n=36, ATLL n=21). Furthermore, we analyzed 57 cases of first progression/relapse. Overall survival (OS) and progression-free survival (PFS) were estimated for newly diagnosed lymphoma (OS1 and PFS1) and at first progression/relapse (OS2 and PFS2), using the Kaplan-Meier method. OS1 and OS2 were further compared by PIT groups. Additionally, possible prognostic factors for OS2 were compared by log rank test. Variables that remained significant in univariate analysis were incorporated in multivariate analysis using the Cox proportional hazard regression model to further investigate independent prognostic factors. For patients who underwent autologous or allogeneic stem cell transplant (SCT), response to therapy was evaluated before SCT.

Results

The overall response rate (ORR) for first line therapy was 74.2% (95% CI: 65.9-81.5%), with a complete response (CR) rate of 58.3% (95% CI: 49.4-66.8%). Median follow up was 26.5 months and the median OS and PFS were 43 months (95% CI: 30-57 months) and 14 months (95% CI: 8-20 months), respectively. There were significant differences in OS1 by subtype; ALK+ ALCL did not achieve the median, and ATLL fared the worst, with median OS1 of 7 months (Figure 1). There was a correlation between PIT scores and OS1 in our cohort (P < .001). Seventy-nine patients (60%) experienced progression/relapse after first line therapy. After excluding patients who did not receive salvage treatment at our institution, we analyzed the remaining 57 cases with progression/relapse (PTCL-NOS n=33, AITL n=10, ALK- ALCL n=4, ALK+ ALCL n=1, NKTCL n=9). The ORR and CR rates at progression/relapse were 46% (95% CI: 32.4-59.3%) and 25% (95% CI: 14.1-37.8%), respectively. The median OS2 was 13 months (95% CI: 6-23 months) and PFS2 was 3 months (95% CI: 2-7 months). There were no cases with ATLL in progression/relapse as all patients with ATLL were either too frail to receive salvage therapy after first progression/relapse, or proceeded to SCT in first remission. Neither PIT at diagnosis nor at first progression/relapse was predictive of OS2. Moreover, there was no correlation between PIT and OS2 in an analysis limited to only patients with PTCL-NOS. Multivariate analysis revealed that presence of B-symptoms (HR: 4.83, 95% CI: 2.10-11.0, P < .001) and Performance Status (PS) greater than 1 (HR: 6.13, 95% CI: 1.84-20.4, P = .003) were significantly associated with poor OS2. Achieving complete or partial remission after first line therapy was associated with superior OS2 (HR 9.34, 95% CI 0.16-0.71, P = .003). Complete or partial remission after first line therapy was the only independent predictive factor for PFS2 (HR 0.31, 95% CI: 0.15-0.65, P = .002).

Conclusion

The presence of B-symptoms and poor PS at progression/relapse were associated with poor OS2. Achieving complete or partial remission after first line therapy was associated with superior OS2. To our knowledge, B-symptoms and response to previous therapy have never been reported as independent prognostic factors. A study with a larger sample size and longer follow-up period is warranted to further assess the validity of these factors.

View largeDownload slide

View largeDownload slide

Close modal