This study will allow the investigators to better assess the efficiency of neoadjuvant chemotherapy before cystectomy by training a predictive model on different patient cohorts with bladder cancer.
The project is based on three prospective cohorts of patients with MIBC: the VESPER trial (n=296), the St-Louis Hospital cohort (n=99), and the COBLAnCE cohort (n=312). Using WES and RNAseq, the investigators will determine genomic instability, DDR gene mutation and molecular subtypes. After digitization of tumour slides, the investigators will train and test predictive models based on deep learning approaches to predict outcome after neoadjuvant chemotherapy, either by estimating molecular subtypes and genetic features from pathological images, or by directly defining a prognostic signature. The statistical analyses will assess the performance of the models combining genomic instability, DNA Damage Response mutations and/or molecular subtyping to predict outcome after neoadjuvant chemotherapy and compare them with the models based on WSI deep learning approaches. These results will help to design new therapeutic strategies.
Study Type
OBSERVATIONAL
Enrollment
707
Blood from patient having muscle invasive bladder cancer who benefit from neoadjuvant chemotherapy with cisplatine
Centre de lutte contre le cancer François Baclesse
Caen, France
Centre de recherche des Cordeliers
Paris, France
Hôpital Saint-Louis AP-HP
Paris, France
Institut Curie Centre de Recherche
Paris, France
Progression-free survival
The time from enrollment to progression or death
Time frame: 3 years
Overall survival
The time from enrollment to death
Time frame: 5 years
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Institut Curie
Paris, France
Institut Gustave Roussy
Paris, France
Mines ParisTech
Paris, France