Abstract
Background: Recent studies (Maurer et al., JCO 2018, Olszewski et al., BJH 2018) have identified short diagnosis-to-treatment interval (DTI) of less than 15 days as a high-risk feature in B-cell lymphomas, independently of other clinical factors. T-cell lymphomas (TCL) pose a challenge because of their heterogeneity, rarity, and poor outcomes. Patients who respond to initial induction chemotherapy frequently proceed to consolidative high-dose chemotherapy with stem cell transplantation (SCT). Our objective was to examine the association between short DTI, receipt of SCT, and survival in 4 most common types of aggressive NK/TCL: peripheral TCL not otherwise specified (PTCL), angioimmunoblastic TCL (AITL), anaplastic large-cell TCL (ALCL), and extranodal NK-/T-cell lymphoma, nasal type (ENKTL).
Methods: From the National Cancer Data Base, a registry containing >80% of lymphomas diagnosed in the US, we selected patients age 18 to 70, diagnosed with PTCL, AITL, ALCL, or ENKTL in 2004-2014, who received multi-agent chemotherapy within 120 days of diagnosis. Short DTI was defined as less than 15 days from diagnosis to start of chemotherapy. We analyzed the use of SCT as part of first course of treatment (before relapse) in a hierarchical logistic model adjusting for age, sex, race/ethnicity, TCL histology, stage, extranodal origin, and type of hospital (academic or community), reporting adjusted odds ratios (OR) with 95% confidence intervals (CI). Overall survival (OS) from start of chemotherapy was examined using Cox models adjusting for the same variables, reporting adjusted hazard ratios (HR). To investigate the differential impact of early mortality, we conducted sensitivity analyses in pre-specified subgroups excluding patients who died within 3 or 6 months of starting chemotherapy, and in the subset who underwent SCT (after presumably favorable response to initial chemotherapy).
Results: Among 8,194 patients (Table; median age 54 years, 38% women), median DTI was 22 days (interquartile range, 11 to 38). The proportion undergoing SCT (11% overall) varied by histology, and was higher in academic centers (17%) compared with other types of hospitals (7%, P<.001). DTI of less than 15 days was associated with a significantly lower likelihood of undergoing SCT overall (OR, 0.71; 95%CI, 0.60-0.85, P=.0002), and this association was present in every histologic subtype except for ALCL, where the use of SCT was altogether rare. The association between short DTI and less frequent SCT persisted in the subset of patients surviving >3 months from the start of chemotherapy (OR, 0.78; 95%CI, 0.65-0.94, P=.012).
With a median follow-up of 10.1 years, survival varied by histology, and was significantly worse for patients with short DTI in PTCL, ALCL and ENKTL, but not in AITL (Fig. A-D). Worse prognosis with short DTI persisted in the subsets patients who survived >3 months (overall HR, 1.32; 95%CI, 1.14-1.33; P<.0001) or >6 months (overall HR, 1.14; 95%CI, 1.04-1.25; P=.005) from the start of chemotherapy. However, in the subgroup of patients who successfully underwent SCT, a short DTI was no longer prognostic (Fig. E; HR, 0.91; 95%CI, 0.66-1.26; P=.57).
Conclusions: Patients with TCL who need to start chemotherapy within 15 days of diagnosis have a lower likelihood of undergoing subsequent SCT. This association is not fully explained by early mortality, and may reflect a more aggressive biology, including worse response to standard chemotherapy-a hypothesis supported by the fact that achievement of SCT nullifies the poor prognostic value of short DTI. Short DTI is thus a novel risk factor in TCL that identifies a high-risk population well suited to experimental treatment strategies, and clinical researchers should strive to enroll these patients. Outcomes from clinical trials in which subjects start therapy after a screening period lasting >15 days from diagnosis should not be uncritically extrapolated to patients who start urgent chemotherapy in general practice. Further research should examine molecular characteristics and response rates to chemotherapy among patients with TCL who require immediate treatment upon diagnosis.
Reagan:Takeda Oncology: Research Funding; Alexion: Honoraria; Pfizer: Research Funding. Olszewski:Genentech: Research Funding; Spectrum Pharmaceuticals: Consultancy, Research Funding; TG Therapeutics: Research Funding.
Author notes
Asterisk with author names denotes non-ASH members.