ctDNA tumor fraction (TF) to predict response to nivolumab in recurrent or metastatic (R/M) head and neck squamous cell carcinoma (HNSCC): An analysis of the multicentric phase 2 TOPNIVO trial.
Abstract
6047 Background: Anti-PD1 provides clinical benefits in HNSCC, yet biomarkers of response remain poorly defined. In other cancer types, measures of ctDNA shedding, such as TF, have been shown to correlate with treatment outcomes. Little is known about ctDNA shedding in HNSCC and its prognostic role. TOPNIVO was a single-arm phase 2 trial designed to assess the tolerability of nivolumab in pretreated R/M HNSCC. The aim of this study is to explore plasma estimation of TF with plasmatic ctDNA and its prognostic role in HNSCC patients receiving nivolumab at 2 nd line or later. Methods: Plasma samples were obtained at baseline in the phase 2 TOPNIVO study (NCT03226756), just before treatment initiation. Genomic copy number alteration and TF were investigated using low-pass Whole Genome Sequencing performed with ctDNA extracted from plasma. Bioinformatic analysis was based on the ichorCNA (V0.2.0). TF > 3% was considered positive. For Tumor Volume (TV) assessment, primary and secondary lesions were analyzed. When available, all lesions were included if fewer than five were present; otherwise, at least five per organ were manually delineated by two experienced physicians. Volume of each lesion were extracted using LIFEx software V.3.44 (Local Image Feature Extraction, www.lifexsoft.org). Patient's TV was then defined by the sum of the volume of all measured lesions. Primary endpoint was overall survival (OS). Results: Plasma samples were available for 86 out of 343 patients, with no major differences compared to the overall cohort in terms of baseline characteristics. In particular, 65 (76%) were male, 12 (14%) had an ECOG PS of 2, 42 (49%) had locally recurrent disease only, and 9 (10%) had HPV-positive oropharyngeal disease. ctDNA TF was positive in all patients (median TF 7.4%, range 4.8% to 35.9%). Median OS of the selected population was 7.4 months (95% CI 5.4 - 11.2). TF was correlated with TV (Spearman 0.31, p = 0.008). Moreover, it was higher in patients with liver metastasis (n = 8, median TF 14% vs 7 %, ) and, for oropharynx, higher in patients with HPV positive disease (n= 9 vs 30, median TF 17% vs 6.7%, ). TF was correlated with OS both in univariate (Hazard Ratio - HR 1.9, p = 0.015 ) and multivariate Cox models (HR 3.12, p = 0.003). Conclusions: In HNSCC patients of the TOPNIVO study, ctDNA TF was always detectable and depends on tumor volume and on the biology of the disease. TF retains independent prognostic validity. More translational data will be presented on biological correlates of ctDNA shed. Multivariable Cox model for overall survival. Variable p HR 95% CI Sex (male vs female) 0.34 1.40 0.70 - 2.83 Age (> 70) 0.87 1.06 0.53 - 2.11 ECOG PS (2 vs 0-1) 0.97 1.02 0.40 - 2.59 Metastatic only disease vs local recurrence 0.028 0.43 0.20 - 0.91 HPV+ oropharynx vs other 0.99 1 0.39 - 2.52 Log TV 0.78 1.03 0.82 - 1.31 Log TF 0.0032 3.12 1.46 - 6.65
Article Details
Journal Info
Journal of Clinical Oncology
Lippincott Williams & Wilkins
Authors (20)
Filippo Gustavo Dall'Olio
Gustave Roussy and Paris-Saclay University, Villejuif, France
Wael Zrafi
Department of Biostatistics and Bioinformatics, Gustave Roussy, Villejuif, France
Adsaya Rathakrishnan
Institut Gustave Roussy, Villejuif, France
Maeva Moreau
Institut Gustave Roussy, Villejuif, France
Rebecca Ibrahim
Institut Gustave Roussy, Villejuif, France
Matthieu Texier
Adrien Laville
Gustave Roussy, Inserm U1030, Villejuif, France
Damien Vasseur
1Gustave Roussy, Villejuif, France
Patrick Saulnier
Institut Gustave Roussy, Villejuif, France
Amaury Daste
Karim Benihoud
Université Paris-Saclay, CNRS, Institut Gustave Roussy, Aspects métaboliques et systémiques de l'oncogénèse pour de nouvelles approches thérapeutiques, 94805, Villejuif, France., Villejuif, France
Pierre Busson
Université Paris-Saclay, CNRS, Institut Gustave Roussy, Aspects métaboliques et systémiques de l'oncogénèse pour de nouvelles approches thérapeutiques, 94805, Villejuif, France., Villejuif, France
Ivan Bieche
Clemence Toullec
Medical Oncology, Institut Sainte Catherine, Avignon, France
Mariana Iacob
Gustave Roussy Cancer Campus Grand Paris, Villejuif, France
Roger Sun
Gustave Roussy, Department of Radiation Oncology, Université Paris-Saclay, UMR 1030, ImmunoRadAI, Villejuif, France
Anne Auperin
Gustave Roussy, Villejuif, France
Laure Monard
R&D Unicancer, Paris, France
Ludovic Lacroix
Caroline Even