Circulating Tumor DNA Predicts Outcomes in Follicular Lymphoma: Analysis from a Prospective Study
Authors
Rahul Lakhotia¹, Stefania Pittaluga², James D. Phelan¹, Allison Graeter³, Christopher Melani¹, Max J.Gordon¹, Jagan R. Muppidi¹, Sam Ng¹, Sumaya Berg¹, Amynah Pradhan¹, Atekelt Tadese¹, Candis Morrison¹, Bonita Bryant¹, Yandan Yang¹, Theresa Davies-Hill², Liza Lindenberg4, Mark A. Ahlman⁵, Esther Mena⁴, Ethan Bergvall⁶, Laura Yee⁷, Nicholas Micheletti⁷, Allison Jacob⁸, Monica Gallucci⁸, Heidi Simmons⁸, Alexander Bagaev⁹, Mark Meerson⁹, Ekaterina Postovalova⁹, Olga Kudryashova⁹, Nikita Kotlov⁹, Nathan Fowler⁹, Elaine S. Jaffe², Louis M. Staudt¹, Wyndham H. Wilson¹, and Mark Roschewski¹
- Lymphoid Malignancies Branch, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD;
- Laboratory of Pathology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD
- Department of Internal Medicine, University of South Florida, Tampa, FL
- Radiology and Imaging Sciences, Clinical Center, National Institutes of Health, Bethesda, MD
- Section of Nuclear Medicine, Medical College of Georgia, Augusta, GA
- Section of Nuclear Medicine, University of Virginia, Charlottesville, VA
- Biostatistics and Data Management Section, Center for Cancer Research, NCI, NIH, Bethesda, MD
- Adaptive Biotechnologies, Seattle, WA
- BostonGene, Corp., Waltham, MA
Abstract
Background:
Follicular lymphoma (FL) shows marked heterogeneity in its clinical course, including spontaneous regression (SR) and histologic transformation (HT). Although watchful waiting (W&W) is routinely employed, monitoring is not standardized. Patients (pts) requiring treatment shortly after diagnosis have worse outcomes, but the biologic basis is unclear and improved classifiers and monitoring tools are needed. Circulating tumor DNA (ctDNA) is a highly tumor-specific biomarker that is prognostic in aggressive B-cell lymphomas. We previously showed that ctDNA in plasma can be identified in >90% FL pts with next-generation sequencing of Ig heavy- and light-chain loci. Here, we present updated results from serial ctDNA monitoring of 78 pts on a prospective clinical trial [NCT03190928].
Methods:
Adult pts with grade 1-3A FL are eligible if they have evaluable disease, no evidence of transformation, and no prior systemic therapy. Pts undergo W&W until they meet uniform protocol-defined treatment criteria and clonal evolution is monitored until 2nd line therapy. Baseline testing includes labs, CT, PET, and research biopsy. Clinic visits are every 4m for 2y, every 6m in years 3-5, then annually. CTs are every other clinic visit. Repeat PET was at 2y, or at suspected disease progression. Cell-stabilizing tubes (plasma) and PBMCs are drawn at each visit. CtDNA analysis was performed in collaboration with Adaptive Biotechnologies as previously described (Distler et al, ASH 2021), and was blinded to clinical outcomes. The primary endpoint was progression requiring treatment within 2y of enrollment. Pts who required treatment within 2y were considered early progressors and those without were considered non-progressors.
Results:
Overall, 78 pts enrolled between July 2017 and Aug 2021, of which 77 had a baseline plasma sample. All 58 (100%) pts with available tumor biopsy had ≥1 dominant clonotype identified in FFPE. In 7 out of 19 pts without FFPE (37%), ≥1 dominant clonotype was identified from plasma . These 65 pts with identifiable clonotype/s constitute the study population. Median age was 57y (range 24-84), including 12 (18%) low-, 25 (39%) intermediate-, and 28 (43%) high-risk by FLIPI. Baseline ctDNA was detectable in 60 pts (92%) with median (interquartile range [IQR]) level of 34 (6-114) counts/mL. Four of 5 pts (80%) with undetectable ctDNA had stage 1-2 FL. Baseline ctDNA levels correlated strongly with FLIPI (P<0.01) and total metabolic tumor volume (TMTV) on PET (P<0.001).
Three pts were unevaluable for progression due to a second cancer, sudden death, and hemolytic anemia. Thirty-three pts (53%) were early progressors and 29 (47%) non-progressors at 2y. Median time to treatment (TTT) was 20m (95% CI, 10-68). Early progressors had a median (IQR) baseline ctDNA level of 38.2 (13-189.4) counts/mL compared to 18.7 (1.2-56.8) counts/mL for non-progressors (P<0.01). Pts with >median baseline ctDNA levels had median TTT of 9.7m (95% CI, 2.8-NE) compared to 37m (95% CI, 17-NE) for pts with
Baseline PBMC clonotype levels were analyzed in 31 pts, which were not associated with FLIPI (P=0.16), TMTV (P=0.49) or treatment within 2y (P=0.65).
Conclusions:
CtDNA is detectable in plasma of >90% pts with untreated FL. Quantitative ctDNA levels correlate with both FLIPI and TMTV. Pts with higher baseline ctDNA and TMTV had a shorter TTT when uniformed treatment criteria were used. Serial ctDNA monitoring shows fluctuating levels that correlate with tumor burden on CT scans, which may provide a non-invasive method to monitor disease without scans. Landmark ctDNA levels and TMTV at baseline do not appear to predict future HT.
Follicular lymphoma (FL) shows marked heterogeneity in its clinical course, including spontaneous regression (SR) and histologic transformation (HT). Although watchful waiting (W&W) is routinely employed, monitoring is not standardized. Patients (pts) requiring treatment shortly after diagnosis have worse outcomes, but the biologic basis is unclear and improved classifiers and monitoring tools are needed. Circulating tumor DNA (ctDNA) is a highly tumor-specific biomarker that is prognostic in aggressive B-cell lymphomas. We previously showed that ctDNA in plasma can be identified in >90% FL pts with next-generation sequencing of Ig heavy- and light-chain loci. Here, we present updated results from serial ctDNA monitoring of 78 pts on a prospective clinical trial [NCT03190928].
Methods:
Adult pts with grade 1-3A FL are eligible if they have evaluable disease, no evidence of transformation, and no prior systemic therapy. Pts undergo W&W until they meet uniform protocol-defined treatment criteria and clonal evolution is monitored until 2nd line therapy. Baseline testing includes labs, CT, PET, and research biopsy. Clinic visits are every 4m for 2y, every 6m in years 3-5, then annually. CTs are every other clinic visit. Repeat PET was at 2y, or at suspected disease progression. Cell-stabilizing tubes (plasma) and PBMCs are drawn at each visit. CtDNA analysis was performed in collaboration with Adaptive Biotechnologies as previously described (Distler et al, ASH 2021), and was blinded to clinical outcomes. The primary endpoint was progression requiring treatment within 2y of enrollment. Pts who required treatment within 2y were considered early progressors and those without were considered non-progressors.
Results:
Overall, 78 pts enrolled between July 2017 and Aug 2021, of which 77 had a baseline plasma sample. All 58 (100%) pts with available tumor biopsy had ≥1 dominant clonotype identified in FFPE. In 7 out of 19 pts without FFPE (37%), ≥1 dominant clonotype was identified from plasma . These 65 pts with identifiable clonotype/s constitute the study population. Median age was 57y (range 24-84), including 12 (18%) low-, 25 (39%) intermediate-, and 28 (43%) high-risk by FLIPI. Baseline ctDNA was detectable in 60 pts (92%) with median (interquartile range [IQR]) level of 34 (6-114) counts/mL. Four of 5 pts (80%) with undetectable ctDNA had stage 1-2 FL. Baseline ctDNA levels correlated strongly with FLIPI (P<0.01) and total metabolic tumor volume (TMTV) on PET (P<0.001).
Three pts were unevaluable for progression due to a second cancer, sudden death, and hemolytic anemia. Thirty-three pts (53%) were early progressors and 29 (47%) non-progressors at 2y. Median time to treatment (TTT) was 20m (95% CI, 10-68). Early progressors had a median (IQR) baseline ctDNA level of 38.2 (13-189.4) counts/mL compared to 18.7 (1.2-56.8) counts/mL for non-progressors (P<0.01). Pts with >median baseline ctDNA levels had median TTT of 9.7m (95% CI, 2.8-NE) compared to 37m (95% CI, 17-NE) for pts with
Baseline PBMC clonotype levels were analyzed in 31 pts, which were not associated with FLIPI (P=0.16), TMTV (P=0.49) or treatment within 2y (P=0.65).
Conclusions:
CtDNA is detectable in plasma of >90% pts with untreated FL. Quantitative ctDNA levels correlate with both FLIPI and TMTV. Pts with higher baseline ctDNA and TMTV had a shorter TTT when uniformed treatment criteria were used. Serial ctDNA monitoring shows fluctuating levels that correlate with tumor burden on CT scans, which may provide a non-invasive method to monitor disease without scans. Landmark ctDNA levels and TMTV at baseline do not appear to predict future HT.
Latest publications
