Next-generation sequencing-based monitoring of circulating tumor DNA reveals clonotypic heterogeneity in untreated PTCL

Abstract

Peripheral T-cell lymphomas (PTCLs) have marked biologic and clinical heterogeneity, which confounds treatment decisions. Advances in circulating tumor DNA (ctDNA) assays using next-generation sequencing (NGS) have improved the detection of molecular relapse and driver mutations in diffuse large B-cell lymphoma and show the potential utility of ctDNA across lymphomas. We investigated NGS-based monitoring of T-cell receptor (TCR) sequences in patients with PTCL undergoing frontline treatment. Of 45 patients, 34 (76%) had tumor-specific clonotypes of the TCRβ or TCRγ genes identified, which included 18 (86%) from baseline tissue and 16 (67%) from baseline serum. Twenty-five (74%) patients had both TCRβ and TCRγ clonotypes, 23 (68%) had more than 1 TCRγ clonotype, and 4 (9%) had multiple TCRβ or TCRγ clonotypes, demonstrating significant intrapatient clonotypic heterogeneity. Among 24 patients with available serial serum samples during treatment, 9 (38%) cleared ctDNA after 2 cycles of therapy, and 11 (46%) had detectable ctDNA at the end of treatment. Patients with detectable ctDNA after therapy showed a trend toward worse survival. Notably, 2 patients with persistently detectable ctDNA after therapy remained in remission with 10 years of follow-up. Clonotypic heterogeneity in tumors and persistence, despite long-term remission, suggests variability in oncological potential. This trial was registered at www.clinicaltrials.gov as #NCT00001337.

© 2021 by The American Society of Hematology. Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0), permitting only noncommercial, nonderivative use with attribution. All other rights reserved.

Figures

Graphical abstract

Figure 1.

Patients and calibration. Fifty-four patients with untreated PTCL enrolled in the treatment study and were screened to determine whether baseline FFPE and/or serum samples were available. Thirty-four patients had a dominant T-cell receptor rearrangement identified from FFPE or serum samples, and they comprised the study cohort for the correlative analysis of circulating tumor DNA.

Figure 2.

Identification of dominant T-cell receptor clonotypes from stored tissue biopsy specimens. (A) The number of dominant clonotypes of TCRγ (mustard) and TCRβ (lavender) identified in both baseline FFPE and serum samples from 18 patients who had at least 1 dominant TCR clonotype identified from baseline FFPE tissue. Of the 18, 16 had matched serum samples, with a dominant clonotype detected in 15. Thirteen dominant TCR clonotypes (lime green) across 7 of those 15 patients were identified in serum that were not identified in FFPE samples. (B) The number of dominant TCRγ (mustard) and TCRβ (lavender) clonotypes detected in baseline serum in 16 patients without available baseline FFPE samples. (C) Tissue source of all dominant TCR clonotypes detected exclusively in FFPE and serum samples, as well as those detected in both tissue sources. (D) The median number and range of clonotypes identified in baseline tissue samples across histologic subtypes of PTCL. ALK− ALCL, ALK− negative ALCL; EATL, enteropathy associated T-cell lymphoma; HSTCL-gd, hepatosplenic T-cell lymphoma gamma delta; PCGDTCL, primary cutaneous gamma delta T-cell lymphoma; PTCL-NOS-L, lymphoepithelioid cell variant of PTCL-NOS; PTCL-TFH, PTCL-T-follicular helper phenotype; SPLTCL, subcutaneous panniculitislike T-cell lymphoma.

Figure 3.

Correlation of ctDNA values with other prognostic parameters. (A) Baseline levels of ctDNA) across IPI categories. (B) Baseline levels of ctDNA correlated with pretreatment serum LDH. ρ, Spearman’s rank correlation coefficient.

Figure 4.

Association between dichotomized ctDNA level and survival. (A) EFS in patients with baseline levels of ctDNA above the median level compared with patients with levels below the median. (B) OS in patients with baseline levels of ctDNA above the median level compared with patients with levels below the median.

Figure 5.

ctDNA kinetics during therapy and clinical outcomes (A) The median and range of quantitative ctDNA levels after each cycle of therapy until treatment is complete. Levels of ctDNA in patients with ALK+ ALCL are depicted with green dots, and levels in other histologic subtypes of PTCL are depicted with black dots. (B) EFS based on presence or absence of ctDNA as a marker of MRD, drawn before day 1 of cycle 3. (C) OS based on the presence or absence of ctDNA as a marker of MRD drawn before day 1 of cycle 3. (D) EFS based on the presence or absence of ctDNA as a marker of MRD at the EoT. (E) OS based on presence or absence of ctDNA as a marker of MRD at EoT. C, cycle.

Figure 6.

ctDNA as surveillance monitoring in clinical progressors. The results of ctDNA detected in stored serum samples collected after therapy compared with conventional monitoring for clinical relapse with periodic clinic visits and regular CT scans. (A) Serum samples collected after therapy in 8 patients who achieved CR, underwent surveillance, and then clinically progressed. The assigned patient numbers are to the left. Histology: patient 9, PTCL-NOS; patients 180, 291, and 317, AITL; patients 190 and 200, ALK+ ALCL; and patients 198 and 274, PTCL-TFH. (B) The pattern of fluctuation of quantitative ctDNA levels of patients 9 and 198 with persistently detectable ctDNA before clinical relapse. The x-axis represents time since the on-study date in months and is different for each graph. The y-axis represents the log10 of the quantity of each dominant clonotype (in clonotypes per milliliter). Each colored line represents a different clonotype. The horizontal gray line represents the median level of baseline ctDNA. The dashed vertical line represents EoT, the red vertical line represents time of progression . The green horizontal bar represents time in clinical remission, and the red horizontal bar represents time after progression.

Figure 7.

Role of ctDNA in surveillance monitoring in nonprogressors. (A) Serum samples collected after therapy in 8 patients who achieved a complete response and underwent surveillance but did not progress clinically. The assigned patient numbers are on the left. Histology: patients 1, 8, and 316, ALK− ALCL; patients 3, 14, and 221, ALK+ALCL; patient 12, PTCL-NOS; and patient 15, AITL. (B) The pattern of fluctuation of quantitative ctDNA levels of patients 1 and 15 with multiple time points of detectable ctDNA after therapy but no overt clinical relapse. The x-axis represents time since the on-study date in months and is different for each graph. The y-axis represents the log10 of the quantity of each dominant clonotype (in clonotypes per milliliter). Each colored line represents a different clonotype. The horizontal gray line represents the median level of baseline ctDNA. The dashed vertical line represents EoT, the red vertical line represents time of progression, and the black vertical line represents end of follow-up. The green horizontal bar represents time in clinical remission.