Molecular subtypes and single tissue and serum biomarkers for the prediction of platinum sensitivity of muscle-invasive bladder cancer
Oláh Csilla
Surgical Medicine Division
Dr. Szijártó Attila
Semmelweis Egyetem Urológiai Klinika
2024-12-02 15:00:00
Urology
Dr. Nyirády Péter
Dr. Szarvas Tibor
Dr. Alpár Donát
Dr. Kuthi Levente
Dr. Takács István
Dr. Borka Katalin
Dr. Buzogány István
Recently, both prognostic and predictive values of distinct molecular BC subtypes have been revealed. In the future, molecular subtypes may play a crucial role in the therapeutic decision-making of patients with MIBC. Current molecular subtyping is based on transcriptome sequencing, which is associated with high demands of input samples, technical complexity, and high costs. This technical barrier largely hindered the widespread adoption of the molecular classification in the daily clinical routine. Therefore, we aimed to develop a simple gene panel-based classification method that can identify molecular subtypes according to TCGA, MDA, LundTax, and Consensus classifications. To this aim, we developed a rule set-based classifier method, which has been applied to fresh-frozen MIBC samples by measuring the gene expression of 68 genes using RT-qPCR. Then, we classified our institutional cohort of 100 MIBC samples into molecular subtypes and examined their prognostic values. In line with the original studies, patients with neuronal subtypes had the worst survival. In addition, we found that strong extracellular matrix and immune cell signatures were associated with improved survival, suggesting that the tumor microenvironment play an important role in tumor progression. As a next step, the rule set-based classifier method was optimized (the marker set was reduced from 68 to 48 genes) and the analytical platform was transferred to the NanoString nCounter technology to enable the analysis of FFPE tumor samples. We then applied our updated rule-set-based classifier to our institutional MIBC cohort, which included adjuvant chemotherapy-treated and non-treated patients. Luminal-papillary subtype (TCGA, Consensus), and urothelial-like subtype (LundTax) were associated with longer survival in the chemo cohort compared to non-chemo cohort. Patients with luminal papillary or urothelial-like subtypes may benefit from adjuvant chemotherapy.
Expression levels of 12 genes with potential chemotherapy predictive value were evaluated in our chemo and non-chemo MIBC cohorts. Patients with high levels of APOBEC3G and CLDN4 as well as with low levels of BIRC5 showed improved survival in the chemo, but not in the non-chemo cohort. These results could be confirmed in an external dataset. In addition, high APOBEC3G expression correlated with the significantly higher pCR rates in patients who received NAC.
High baseline serum levels of SDC1 and MMP7 were associated with inferior survival outcomes in MIBC patients who received platinum-based chemotherapy
