Circulating KRAS G12D but not G12V is associated with survival in metastatic pancreatic ductal adenocarcinoma

Jacob E. Till, Lee McDaniel, Changgee Chang, Qi Long, Shannon M. Pfeiffer, Jaclyn P. Lyman, Lacey Padrón, Deena M. Maurer, Jia Xin Yu, Christine N. Spencer, Pier Federico Gherardini, Diane M. Da Silva, Theresa LaVallee, Charles W. Abbott, Richard Chen, Sean Michael Boyle, Neha Bhagwat, Samuele Cannas, Hersh Sagreiya, Wenrui Li, Stephanie S. Yee, Aseel Abdalla, Zhuoyang Wang, Melinda Yin, Dominique Ballinger, Paul Wissel, Jennifer R. Eads, Thomas B. Karasic, Charles L. Schneider, Peter J. O’Dwyer, Ursina Teitelbaum, Kim A. Reiss, Osama E. Rahma, George A. Fisher, Andrew H. Ko, Zev A. Wainberg, Robert A. Wolff, Eileen M. O’Reilly, Mark H. O’Hara, Christopher R. Cabanski, Robert H. Vonderheide, Erica L. CarpenterNature Communications 2024 · DOI

Abstract

While high circulating tumor DNA (ctDNA) levels are associated with poor survival for multiple cancers, variant-specific differences in the association of ctDNA levels and survival have not been examined. Here we investigate KRAS ctDNA (ctKRAS) variant-specific associations with overall and progression-free survival (OS/PFS) in first-line metastatic pancreatic ductal adenocarcinoma (mPDAC) for patients receiving chemoimmunotherapy ("PRINCE", NCT03214250), and an independent cohort receiving standard of care (SOC) chemotherapy. For PRINCE, higher baseline plasma levels are associated with worse OS for ctKRAS G12D (log-rank p = 0.0010) but not G12V (p = 0.7101), even with adjustment for clinical covariates. Early, on-therapy clearance of G12D (p = 0.0002), but not G12V (p = 0.4058), strongly associates with OS for PRINCE. Similar results are obtained for the SOC cohort, and for PFS in both cohorts. These results suggest ctKRAS G12D but not G12V as a promising prognostic biomarker for mPDAC and that G12D clearance could also serve as an early biomarker of response.

Introduction

Metastatic pancreatic ductal adenocarcinoma (mPDAC) is characterized by high mortality and significant challenges in disease monitoring. Traditional tissue biopsies are often invasive or yield insufficient material, while standard blood markers like CA19-9 are unreliable in a subset of patients. Circulating tumor DNA (ctDNA) has emerged as a non-invasive "liquid biopsy" to track tumor burden.

While KRAS mutations are present in approximately 90% of pancreatic cancers, clinical research often treats these mutations as a monolithic indicator. This study challenges that assumption by investigating whether the two most common KRAS variants, G12D and G12V, provide distinct prognostic information regarding patient survival and treatment response.

Methods and Study Design

The researchers utilized plasma samples from two distinct patient groups to ensure the robustness of their findings. The primary cohort consisted of 129 patients from the PRINCE Trial (NCT03214250) receiving chemoimmunotherapy. An independent SOC Cohort of 85 therapy-naive patients receiving standard-of-care chemotherapy was used for validation.

Flowchart of patient enrollment and sample collection for the PRINCE and SOC cohorts.

Mutation detection and quantification were performed using Droplet Digital PCR (ddPCR), a highly sensitive technique with a limit of detection of approximately 0.04%. The primary metric was the Variant Allele Fraction (VAF)—the percentage of circulating DNA that is mutant—measured at baseline and eight weeks into therapy to assess "clearance."

Key Results

Variant-Specific Prognosis

The study revealed a striking divergence between the two KRAS variants. In the PRINCE trial, patients with high baseline G12D levels had a median overall survival (OS) of only 6.4 months, compared to 18.1 months for those with low levels (p=0.0010). In contrast, there was no significant difference in survival based on G12V levels (p=0.7101).

Kaplan-Meier survival curves for Overall Survival (A) and Progression-Free Survival (C) stratified by median Variant Allele Fraction (VAF).

On-Therapy Clearance

Early clearance of ctDNA (becoming undetectable by week 8) was a powerful predictor for G12D patients. Those who cleared G12D lived a median of 19.0 months, while those who did not lived only 7.6 months (p=0.0002). Again, G12V clearance showed no significant association with survival (p=0.4058). These results were consistent across both the immunotherapy and chemotherapy cohorts, suggesting that the prognostic value of G12D is a fundamental property of the disease biology.

Discussion and Limitations

The findings suggest that G12D levels correlate more closely with imaging-based tumor burden than G12V levels, potentially reflecting differences in how these tumors shed DNA or their underlying biological aggressiveness. This research paves the way for personalized liquid biopsy reports that provide nuanced risk assessments based on specific mutation variants.

However, the study acknowledges limitations, including smaller sample sizes for the G12V subgroup and a lack of racial diversity in the cohorts. Future work is needed to elucidate the biological mechanisms behind these variant-specific differences and to validate the findings in broader, more diverse populations.

Summary updated

Abstract

While high circulating tumor DNA (ctDNA) levels are associated with poor survival for multiple cancers, variant-specific differences in the association of ctDNA levels and survival have not been examined. Here we investigate KRAS ctDNA (ctKRAS) variant-specific associations with overall and progression-free survival (OS/PFS) in first-line metastatic pancreatic ductal adenocarcinoma (mPDAC) for patients receiving chemoimmunotherapy ("PRINCE", NCT03214250), and an independent cohort receiving standard of care (SOC) chemotherapy. For PRINCE, higher baseline plasma levels are associated with worse OS for ctKRAS G12D (log-rank p = 0.0010) but not G12V (p = 0.7101), even with adjustment for clinical covariates. Early, on-therapy clearance of G12D (p = 0.0002), but not G12V (p = 0.4058), strongly associates with OS for PRINCE. Similar results are obtained for the SOC cohort, and for PFS in both cohorts. These results suggest ctKRAS G12D but not G12V as a promising prognostic biomarker for mPDAC and that G12D clearance could also serve as an early biomarker of response.
BibTeX
@inproceedings{till2024circulating,
  author = {Jacob E. Till and Lee McDaniel and Changgee Chang and Qi Long and Shannon M. Pfeiffer and Jaclyn P. Lyman and Lacey Padrón and Deena M. Maurer and Jia Xin Yu and Christine N. Spencer and Pier Federico Gherardini and Diane M. Da Silva and Theresa LaVallee and Charles W. Abbott and Richard Chen and Sean Michael Boyle and Neha Bhagwat and Samuele Cannas and Hersh Sagreiya and Wenrui Li and Stephanie S. Yee and Aseel Abdalla and Zhuoyang Wang and Melinda Yin and Dominique Ballinger and Paul Wissel and Jennifer R. Eads and Thomas B. Karasic and Charles L. Schneider and Peter J. O’Dwyer and Ursina Teitelbaum and Kim A. Reiss and Osama E. Rahma and George A. Fisher and Andrew H. Ko and Zev A. Wainberg and Robert A. Wolff and Eileen M. O’Reilly and Mark H. O’Hara and Christopher R. Cabanski and Robert H. Vonderheide and Erica L. Carpenter},
  title = {{Circulating KRAS G12D but not G12V is associated with survival in metastatic pancreatic ductal adenocarcinoma}},
  booktitle = {Nature Communications},
  year = {2024},
  doi = {10.1038/s41467-024-49915-5},
}
Figure