Michael C. Jin1, David M. Kurtz1, Mohammad Shahrokh Esfahani1, Brian J. Sworder1, Joseph Schroer-Martin1, Joanne Soo1, Cynthia Glover1, Mark Roschewski2, Wyndham Wilson2, Ulrich Düehrsen3, Andreas Hüttmann3, Davide Rossi4, Gianluca Gaidano4, Jason Westin5, Maximilian Diehn1, Ranjana Advani1, Ash A. Alizadeh1
1 Division of Oncology, Department of Medicine, Stanford University, Stanford, California, USA, 2 Lymphoid Malignancies Branch, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA, 3 Clinic for Hematology, University Hospital Essen, Essen, Germany, 4 Division of Hematology, Department of Translational Medicine, University of Eastern Piedmont, Novara, Italy, 5 Department of Lymphoma/Myeloma, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
Introduction: Circulating tumor DNA (ctDNA) analysis has an emerging diagnostic role in multiple malignancies including lymphomas. In classical Hodgkin Lymphoma (cHL), where malignant Reed Sternberg (RS) cells are rare and typically require microdissection from archival tissues, ctDNA could enable noninvasive genotyping of somatic single nucleotide variants (SNVs) and somatic copy number alterations (SCNAs).
Methods: 24 subjects with cHL from Stanford were studied, including 16 (67%) early stage and 8 (33%) advanced disease. Plasma samples were sequenced with CAPP-Seq (Newman et al Nat Biotech 2016), using a panel informed by tumor biopsies. cHL patients were compared to previously analyzed subjects, including 121 DLBCL NOS and 22 PMBCL. Given the thoracic distribution of most cHL, we also compared ctDNA levels to 55 lung carcinomas (NSCLC). Additional subjects are currently being profiled and will be presented at ISHL.
Results: The median pretreatment ctDNA level in cHL was 212 hGE/mL (22–1918), corresponding to a median variant allelic level of 2.8% (0.3–13.6) (Fig 1A). Pretreatment ctDNA levels in cHL were significantly correlated with total metabolic tumor volume (MTV) (Spearman = 0.615, p < 0.01) (Fig 1B) but not with other common clinical characteristics. Surprisingly, despite the lower tumor purity of RS cells in cHL than malignant B-cells in DLBCL, the relationship between ctDNA and PET/CT MTV in cHL was highly similar to that of DLBCL. Specifically, cHL and DLBCL were indistinguishable for the ratio between ctDNA and MTV (mean ctDNA/MTV of 2.1 vs 1.5 hGE/mL per cm^3 tumor, p = NS) and both were significantly higher than that of NSCLC (Fig 1C).
Using our method for detecting SCNAs in ctDNA, we noninvasively genotyped PD-L1 gains in 47% of cHL patients, significantly more frequently observed than in DLBCL patients (p = 0.02) (Fig 1D) (Jin et al ASH 2017). When assessing minimal residual disease (MRD) detection in simulated cHL cfDNA samples of various purities, we found a lower detection limit of 0.0046%, with 100% sensitivity at ctDNA levels above 0.01%.
Conclusions: ctDNA levels in cHL are higher than expected based on tumor purity, with pre-treatment levels similar to DLBCL, and allows for reliable genotyping of cHL at diagnosis or relapse. Assessment of MRD with ctDNA in cHL demonstrates a remarkably low detection limit. These findings support the role of ctDNA as a noninvasive biomarker for cHL patients.