Recent publications of this group

In order to gain increased understanding of the role of bacteria in cancers, we used a pan-cancer cohort of over 4000 metastatic tumors to create a comprehensive catalogue of the microbial community present in metastatic cancers. We showed that the composition of the tumor microbiome is largely dependent on the anatomical location of the tumor and find an enrichment of anaerobic bacteria in hypoxic tumors. Additionally, we find new associations between microbial diversity and immune activation and observed that the presence of intratumoral Fusobacterium is associated with poor response to immune checkpoint blockade in NSCLC patients. Taken together, we created a large resource of the metastatic tumor microbiome, which may help the development of novel microbiome-targeting treatment strategies.

Recently, our perspective about generating and using real-world data in oncology has been accepted by Cell. In this perspective, we highlight the challenges that the current precision oncology paradigm has revealed in the feasibility and data generalizability of traditional clinical trials and go over the potential of real-world data to fill the resulting gap in research methodology. Despite prevailing concerns regarding data quality and comprehensiveness, privacy and bias, real-world data may play a pivotal role in supplementing clinical trials (enabling conditional reimbursement and accelerated drug access) and innovating trial conduct. Additionally, real-world data can serve as a valuable resource for biomarker discovery and validation.

We recently published results of a collaborative effort studying the value of whole genome sequencing for precision immunotherapy in non-small cell lung cancer (NSCLC). In recent years, PD-1 blockers have transformed the standard of care for this disease, showing impressive results in some patients. Yet, the majority of patients with NSCLC do not respond to these treatments and hence suffer from overtreatment, resulting in unnecessary toxicity for patients and high costs for our health care systems. To address these challenges, we studied 254 patients with NSCLC to determine if a comprehensive DNA test could be used to identify the resistant patients already before treatment was initiated. Here, we successfully discovered and validated a combinatorial biomarker strategy to identify a large subgroup of patients with ultralow response rates, which has clear potential to guide treatment for this disease.