ctDNA and Fragmentomics Reveal New Predictors of Survival in 177 Lu-PSMA–617 mCRPC

Metastatic castration-resistant prostate cancer (mCRPC) remains one of the most challenging stages of prostate cancer, marked by poor prognosis and limited durable treatment options. The radioligand therapy 177Lu-PSMA-617 (Pluvicto) has emerged as a significant advance, demonstrating an overall survival benefit in selected patients. However, clinical responses vary widely, highlighting an urgent need for biomarkers that can better predict which patients will benefit and why resistance occurs.

A new real-world prospective study involving 140 patients provides compelling evidence that circulating tumor DNA (ctDNA) analysis may offer a powerful, non-invasive solution to this problem. By leveraging whole genome sequencing of liquid biopsies collected before and during treatment, researchers were able to quantify ctDNA fraction and characterize tumor-derived genomic features, including copy number alterations, somatic mutations, and structural rearrangements. In parallel, a novel computational framework known as Proteus was used to extract fragmentomic signals from ctDNA, enabling inference of gene expression patterns and tumor phenotypes such as proliferation, hypoxia, and lineage plasticity.

One of the most striking findings is the strong prognostic value of baseline ctDNA fraction. Patients with higher ctDNA levels prior to treatment had significantly worse overall survival, with a hazard ratio of 2.19, independent of PSA response. This suggests that ctDNA fraction captures aspects of disease burden and biology that PSA alone cannot fully reflect. Consistent with this, ctDNA levels correlated closely with total tumor volume measured by both PSMA-PET and FDG-PET imaging, reinforcing its role as a surrogate for systemic disease extent.

Beyond tumor burden, the study highlights the importance of underlying tumor biology. Patients with poor outcomes exhibited higher genomic instability and elevated tumor mutational burden, indicating a more aggressive disease phenotype. Specific genomic alterations further refined prognostic stratification. Deleterious mutations in the DNA repair gene ATM were associated with improved survival, potentially reflecting increased sensitivity to radioligand-induced DNA damage. In contrast, amplification of the oncogene MYC was linked to inferior outcomes, consistent with its role in driving proliferation and treatment resistance.

Taken together, these findings support a shift toward a multi-layered biomarker framework that combines ctDNA quantity, genomic alterations, and inferred tumor phenotypes. Such an approach could improve patient selection for 177Lu-PSMA therapy, enable earlier identification of non-responders, and guide the development of combination strategies aimed at overcoming resistance. In clinical practice, this may translate into more personalized treatment pathways, where patients with high-risk molecular profiles are directed toward intensified or alternative therapeutic approaches rather than standard radioligand therapy alone.

Source.