Dados do Resumo

Título

Is the Intratumoral genetic heterogeneity of Locally Advanced Rectal Cancer observed at the level of gene expression?

Introdução

Locally advanced rectal tumors (LARC) are treated with neoadjuvant chemoradiotherapy (nCRT) followed by total mesorectal excision. Nearly 50% of patients achieve a complete clinical response (cCR) post-nCRT allowing organ preservation to be established. However, stratification based on clinical-radiological criteria has mostly failed to predict regrowth. Our group has demonstrated that LARC exhibit high intratumoral genetic heterogeneity (ITGH), with increased levels in non-responders post-nCRT.

Objetivo

To investigate ITGH as potential biomarker of nCRT response, measuring it from genomic and transcriptomic data of single biopsies. This investigation seeks to improve patient stratification and reduce risks such as local recurrence and metastasis in non-responders.

Métodos

We collected genomic (MSK-IMPACT plataform and whole exome sequence - WES), clinical (cBioportal) and transcriptomic (GEO, GSE209746) data from the Chatila et al. (2022) study.
MATH scores were calculated using somatic mutation calls and their corresponding mutant allele frequencies (MAFs) with statistical packages and the specific maftools library in R (v. 4.4.1) environment.
DEPTH2 scores ITGH was calculated by taking the standard deviation of the absolute z-scores from the normalized gene expression matrix. We primary used the authors' FPKM-normalized data, which were likewise processed in R.
These results were evaluated in relation to cCR or pathological complete response (pCR). Patients followed-up in two years were stratified in nCRT-responders (nCRT-R) and non-responder (nCRT-NR) groups.

Resultados

An initial dataset of 357 samples from 266 patients was analyzed, but MATH scores were available for 232 patients due to the limited number of mutations. This final group consisted of 100 women and 132 men, with a mean age of 54.3 years.
The total number of mutations was 17,859, with median MATH scores of 42.12 in nCRT-R (N=168) versus 36.04 in nCRT-NR (N=64), indicating a tendency toward higher ITGH values in nCRT-R, though not statistically significant (p = 0.18, Mann-Whitney test). These ITGH measurements, however, were obtained using four different gene panels for 168 patients, with only 48 patients evaluated by WES, where the MATH score has been previously applied.
ITGH measured from expression data also revealed higher values in the nCRT-R, with a median of 0.695 compared to 0.616 in the nCRT-NR, with a statistically significant p-value of 0.045 (Mann-Whitney test). We intend to test the RSEM expression data normalization method to corroborate these preliminary findings.

Conclusões

This is the first time ITGH has been simultaneously measured from genomic and expression data of LARC samples, and these preliminary results provide insights for future studies in our local cohorts. We aim to conclude this study by addressing questions such as: Can we accurately estimate MATH scores from panel data? Are ITGH measurements from genomic and transcriptomic data comparable? In addition to nCRT response, can ITGH be associated with prognosis?

Palavras Chave

LARC; intratumoral genetic heterogeneity; prognostic biomarker