Introduction:

The prevailing thinking about PDAC is that primary and metastatic lesions have similar frequencies of genetic mutations. In addition, metastatic lesions are thought to have the same driver mutations no matter which organ they occupy. The majority of the data in this area are drawn from studies which examine either metastatic lesions alone or in tandem with unpaired analyses of primary tumors. By comparing the molecular profiles of paired patient samples, we hope to further investigate the mutations involved in metastasis of PDAC.

Background:

Pancreatic cancer is the third leading cause of cancer-related death in the U.S. The vast majority of pancreatic cancers are PDAC (90%). Most patients with PDAC die from complications from met disease, thus it is vital to better understand the molecular changes that promote met spread. Prior studies have shown few molecular changes between primary and met PDAC in unpaired analyses, but limited data exist on large-scale comparisons between paired samples from the same patient.

Methods:

We analyzed next-generation sequencing (NGS) and immunohistochemistry (IHC) data from 123 patients with multiple PDAC tumors profiled by Caris Life Sciences and compared paired primary and met samples from the same patient. After patients with unrelated second primary cancers were excluded, 113 pairs were used for analysis. McNemar’s test was used to compare primary and met tumor pairs.Results: The most common sites of mets were liver (33%), lung (19%), and peritoneum (18%). The average time between samples was 21.3 months (range 1-92). KRAS status changed in 13.2% of pairs (5.7% gained, 7.5% lost, n = 53, P = 0.710), TP53 status changed in 16.1% (16.1% gained, 0% lost, n = 31, P = 0.025), and SMAD4 status changed in 14.8% (7.4% gained, 7.4% lost, n = 27, P = 1.000). Mets gained expression of TOPO1 (37.5%, n = 80, P = 0.003), TOP2A (42.6%, n = 47, P = 0.003), PTEN (27.3%, n = 66, P = 0.050), and PD-L1 (11.1%, n = 36, P = 0.180). Tumor mutational burden (TMB) increased in mets in 11 of 12 pairs with TMB data (mean 6.67 v. 4.42 mutations per megabase, P = 0.0015 by Wilcoxon signed-rank test).

Conclusion:

KRAS mutational status between primary and met PDAC pairs was usually concordant but changed more often than previously reported. TOPO1, TOP2A, and PTEN expression were significantly discordant. The rate of discordance and increase in TMB support profiling of PDAC mets. Continued research into which mutations play key roles in PDAC mets could yield new targets for future therapies.