Investigations reveal significant differences between patient-derived xenografts and original ovarian tumors through proteogenomic analysis.
Researchers utilizing patient-derived xenograft (PDX) models are challenging long-held assumptions about their effectiveness for studying ovarian cancer. A recent study highlights significant discrepancies between the proteogenomic profiles of serially passaged PDXs and their patient tumors of origin. These findings could reshape how the scientific community approaches ovarian cancer research and patient-specific treatments.
Ovarian cancer, one of the most lethal gynecological cancers, presents complex challenges due to its various subtypes, with high-grade serous carcinoma (HGSC) being the most prevalent. Although PDX models are commonly used to study tumor behavior and drug responses, evidence has emerged indicating they may not faithfully represent the original patient tumors due to significant transcriptional differences.
Utilizing quantitative mass spectrometry alongside RNA sequencing, researchers have exposed distinctive proteomic divergences across successive passages of PDXs derived from ovarian cancer patients. They established patient-specific databases to guide their analysis, identifying over 7,000 proteins within the original tumors, resulting in more accurate representations of the human proteome compared to conventional databases.
"Our findings highlight features of distinct and dynamic proteomes of serially-passaged PDX models of ovarian cancer," the authors noted, underscoring the discrepancies observed between tumor models and the originating patient tumor. Indeed, their analysis shows not only different quantities of proteins across various PDX passages but also distinct biological processes involved like extracellular matrix organization and immune responses.
These limitations have been emphasized with the observation of unstable protein levels across passaged models. Specifically, proteins associated with cancer-related pathways displayed significant fluctuations as they were serially passaged through mice. The study articulates how the initial passages actively altered the proteomic landscapes, diverging significantly from the patients’ own tumors.
Importantly, when PDX models are subjected to successive passages, the unique characteristics of the original tumors become less identifiable. The research indicates this loss can undermine the reliability of such models for translational studies and clinical applications. "Protein abundances cluster independently of the patient of origin and tumor passaging," the study concludes.
These revelations stress the necessity for utilizing more refined methodologies, emphasizing integrative approaches like proteogenomics to ascertain the validity of PDX models. The utilization of patient-specific databases is proving advantageous, as this approach aids researchers to retain the unique genetic features of patient tumors even through serial passaging.
Despite the challenges these findings present, they may lead to significant improvements in future ovarian cancer research. Acknowledging the dynamic nature of tumor biology during the creation and use of PDX models may guide the development of more effective treatment strategies and models enhancing translational research efforts.
The study reveals urgent questions concerning the stability of important cancer markers and encourages future investigations to minimize discrepancies between PDXs and patient tumors.
The changes witnessed within the tumor environment during PDX engagements call for optimized methods to maintain the integrity of original cancer specimens throughout preclinical experimentation.
Moving forward, maintaining the connection between patient-derived tumors and PDX models will be integral for achieving advancements within personalized medicine. With cancer treatment becoming increasingly focused on customization, ensuring accuracy from PDX models will remain pivotal for fostering scientific discovery and therapeutic progress.