Abstract 4136: Tracking metastatic dissemination and tumor growth using longitudinal imaging and ctDNA

Abstract

Abstract Metastatic lung cancer is a heterogeneous disease, characterized by different patterns of clinical progression and therapy response. The scale of this heterogeneity on a lesion specific level has yet to be quantified but is of clinical importance because it may provide an indication of molecular mechanisms underpinning metastasis growth. Currently, whether metastasis growth patterns can be predicted by the primary tumor or governed by metastasis specific characteristics is not known. Furthermore, circulating tumor DNA (ctDNA) has proven useful to predict disease relapse, but how closely it reflects metastatic growth post relapse is unclear. In 104 patients enrolled in the TRACERx study with disease relapse post-surgical resection of the primary tumor, lesion volumetric dynamics were tracked in 460 metastases. Total volume growth rate (TVGR) was calculated using CT imaging performed between disease relapse and death. All patients had whole exome sequencing (WES) of the primary tumor. For 19 patients enrolled in the PEACE autopsy study WES and RNAseq from 60 metastases visible on imaging prior to death and subsequently sampled at autopsy were analyses. In 29 patients, ctDNA was used to track 200 tumor-specific mutations to assess tumor burden and detect metastatic subclones. High TVGR correlated with poor survival (hazard ratio for death 1st vs 2nd tertile was 0.21, p<0.001; 2nd vs 3rd tertile was 0.51, p=0.04). Primary tumor subclonal whole genome doubling was predictive of high TVGR. There was considerable heterogeneity in the growth rate of metastases within patients. The anatomical location of metastases contributed to this, extrathoracic soft tissue and pleura had the highest growth rates, whilst lymph node and adrenal the lowest (p<0.001). Molecular attributes differed between metastases, with fast growing metastases being enriched for proliferation pathway gene expression. It was observed that the most prevalent population of cells, or clone, in the fastest growing metastasis within a patient was exclusive to that metastasis, suggesting that genomic features that characterize the clone might support rapid growth. ctDNA tracked with tumor burden post relapse and patients with extrathoracic relapse had higher levels of ctDNA. In 9 patients, metastasis specific mutations identified at autopsy and in ctDNA, were used to determine the proportion of ctDNA shed by different metastatic sites. ctDNA subclonal fraction was found to track with changes in lesion volumes during disease progression. This work highlights significant heterogeneity in metastatic growth rates which can potentially be attributed to the presence of distinct genomic features which may dictate tumor growth rate. ctDNA accurately reproduces patterns of radiological disease progression and when combined with longitudinal imaging and tumor DNA sequencing, can accurately reconstruct the natural history of cancer evolution. Citation Format: Wing Kin Liu, Boyue Ding, Hessey Sonya, Carlos Martinez-Ruiz, Cristina Naceur-Lombardelli, Corentin Richard, Hyothaek Lee, Catarina Veiga, Kishen Patel, Ariana Huebner, Allan Hackshaw, Christopher Abbosh, Alexander M. Frankell, Gary Royle, Charles Swanton, Mariam Jamal-Hanjani. Tracking metastatic dissemination and tumor growth using longitudinal imaging and ctDNA [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4136.