Spatial biomarkers of response to neoadjuvant therapy in muscle-invasive bladder cancer: the DUTRENEO trial
Grande, E.; Sibai, M.; Andrada, E.; Grases, D.; Reig, O.; Escobosa, M.; Perez, E.; Marques, M.; Martinez-Villarreal, J.; Benitez, R.; Malats, N.; Moreno, A.; Garcia, J.; Galvan, P.; Prat, A.; Azueta, A.; Castellano, D.; Guerrero, F.; Puente, J.; Galante, M. I.; Font, A.; Garcia Del Muro, F. J.; Alonso, T.; Burgos, J.; Pinto, A.; Alvarez, M.; Climent, M. A.; Paramio, J.; Maroto, P.; Dominguez, M.; Duran, I.; Porta-Pardo, E.; Real, F. X.
Show abstract
Many studies have reported biomarkers predictive of response to immune checkpoint inhibitors (ICI) based on retrospective analyses. However, few clinical trials have tested their value prospectively. The DUTRENEO trial (EudraCT: 2017-002246-6) investigated whether an 18-gene Tumour Inflammation Signature (TIS) that can robustly identify patients who respond to ICI in multiple tumour types could stratify patients with localized muscle-invasive bladder cancer (MIBC) to receive neoadjuvant ICI (durvalumab+tremelimumab) or standard cisplatin-based neoadjuvant chemotherapy (NAC). Patients with TIS-high tumors were randomized to ICI or NAC, while patients with TIS-low tumors received NAC. A total of 73 patients were treated. Pathological complete response (pCR) rates were 38.5% (TIS-High, ICI), 30% (TIS-High, NAC), and 55% (TIS-Low, NAC) (p = 0.349), indicating that - as applied - the TIS score did not significantly enrich in responders to ICI. Post-hoc analysis showed that higher TIS thresholds improved prediction of response to ICI but excluded many responders. Multi-omics analyses of pre-treatment samples, including whole-exome sequencing, bulk RNA sequencing, and spatial transcriptomics (Visium, Xenium), revealed that bulk RNA response signatures originated mainly from cancer cells. Spatial transcriptomics showed that ICI response was associated with proximity between cancer cells and adaptive immune cells, while resistance to NAC was linked to phenotypic plasticity of cancer cells despite low genetic diversity. The unique design of the DUTRENEO trial underscores the challenges of translating retrospective biomarkers into clinical practice and highlights the importance of spatial features in understanding tumour-immune interactions. These findings suggest that integrating spatial and multi-modal biomarkers in well-designed clinical trials could improve stratification and response prediction to select neoadjuvant therapy.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.