Novel Single Agents Are Equivalent to Conventional Chemotherapy Inpatients with Relapsed and Refractory Mature T-Cell Lymphomas: A Meta-Analysis

Background: Patients with advanced peripheral and cutaneous T-cell lymphomas (PTCL and CTCL) have an unfavorable prognosis . Primary refractoriness to traditional chemotherapy and relapse is common. Few FDA-approved "novel" single agents (SA) are available but their effectiveness as salvage agents relative to traditional multiagent cytotoxic chemotherapy (CC) regimens remains unknown. Therefore, we conducted a systematic meta-analysis to compare the response rates of approved and experimental single agents to conventional chemotherapy for patients with relapsed and refractory (R/R) PTCL and CTCL.

Methods: This systematic review is reported in accordance with the PRISMA guidelines. Briefly a Harvard University Librarian (PB) systematically searched the publication libraries of: MEDLINE (through Ovid SP), Embase, Web of Science Core Collection, and Cochrane's Central Register for adult patients with R/R PTCL and CTCL enrolled in phase I, II, and III clinical trials of novel single and cytotoxic drugs. Agents determined as novel included antifolates, histone deacetylase inhibitors (HDACi), antibody-drug conjugates, therapeutic antibodies, hypomethylating agents, cereblon modulators, inhibitors of PI3K/AKT/mTOR, JAK/STAT, aurora kinase, farnesyl transferase and proteosome pathways, and CD4 CAR T- cells. Conventional chemotherapy agents included ifosfamide-, gemcitabine-, anthracycline-, and platinum-based treatments.

Three researchers (NS, RS and LB) independently reviewed eligible studies and extracted data for baseline demographics, histological subtype, treatment characteristics including response rates, duration of response, overall and progression free survival, toxicity, and risk of bias assessment. Primary outcome was overall response rate (ORR), defined as the best reported partial response (PR) or better. Meta-analyses were conducted for ORR and the scale of logarithm of odd using the generic inverse variance method. The random effects model was used to pool the effect sizes for each study assuming a normal distribution of the random effects. Preplanned subgroup by therapy type and histological subtypes of lymphoma were treated as fixed effects and compared using Wald test. The degree of statistical heterogeneity was evaluated by inconsistency index (I2).

Results:

Our literature search identified 1873 articles, which after filtering for inclusion and exclusion criteria, ruling in only papers including T-cell lymphoma specific response, was narrowed to 128 studies for data extraction. Of these 128 studies, 35 were phase I trials, 70 phase II, 13 phase III, and 10 were combination. First, we divided all studies into upfront (n=33) versus R/R (n=100) based on timing of their treatments. We specifically focused on patients with R/R TCLs and subdivided those studies into the novel SA (n=84) and conventional chemotherapy (n=16) categories. The ORR for novel agents was lower at 37% (95% confidence interval [CI]: 34, 41) in comparison with 55% (95% CI: 40, 69) for standard chemotherapy agents (p=0.02).

When stratified by histological subtype, patients with PTCL-NOS (n=751) had comparable ORR to novel agents (31%; 95% CI: 27, 35) and conventional chemotherapy (40%; 95% CI: 31,50; p=0.08). Similar results were seen with patients with AITL (n=296) who achieved equivocal ORR to single agents (45%; 95% CI: 38, 52) when contrasted with conventional chemotherapy (55%; 95% CI: 27, 80; p=0.52). Similar efficacy was seen for patients with CTCL (n=612) across SA (34%, 95% CI:29, 40) and CC (44%; 95% CI: 33, 56; p=0.11).

Conclusions: Our meta-analysis highlights that for particular histological subtypes of PTCL such as PTCL-NOS and AITL, single agents are non-inferior to cytotoxic chemotherapy regimens in the R/R setting. These findings warrant serious consideration in the design and development of clinical trials for patients with R/R PTCL and the need for tailored treatments. Our meta-analysis also informs clinicians and patients about the benefits of single agents in comparison with standard chemotherapy while making clinical decisions for specified histologies.