First-in-Human, Phase I Study of PCA062 in Solid Tumors—Letter

In a recent issue, Duca and colleagues reported the results of a first-in-human phase I study of PCA062, an antibody–drug conjugate (ADC) targeting P-cadherin (CDH3), in patients with CDH3-positive solid tumors (1). The clinical development of PCA062 was, however, terminated due to its limited antitumor activity at MTD. Here, we would like to point out that the therapeutic development of a third-generation CDH3-ADC should still be considered as an interesting strategy for pancreatic adenocarcinoma (PAAD).PAAD is one of the most pejorative cancers with few treatment options. The most represented tumors in the Duca and colleagues’ study were esophagus and head and neck squamous cell carcinomas (respectively, ESCC n = 9, and HNSCC n = 6); the detailed information on the tumor type was not available for the other 31 patients enrolled, impeding to identify the number of patients with PAAD (at least one). Although inferior to expression in the most represented tumors in Duca and colleagues’ study, PAAD samples expressed moderate levels of CDH3, with very elevated fold-changes when compared with normal tissues (Fig. 1A). In our database (1,090 PAAD tumors), we confirmed the significant upregulation of CDH3 in primary tumors and metastases (Fig. 1B), and CDH3 overexpression was an independent poor-prognosis factor for survival (Fig. 1C). The negative Duca and colleagues’ results, notably in tumors with high CDH3 expression (ESCC, HNSCC), did not suggest that CDH3 expression could be a predictive marker for response to CDH3-ADC. A low ABCC1/SLC46A3 gene expression ratio was associated with response to this CDH3-ADC in preclinical models (2); interestingly, we found that PAAD displayed the smallest ratio when compared with 18 other carcinoma types, including HNSCC and ESCC (that displayed high and unfavorable ratios; Fig. 1D), suggesting higher potential vulnerability of PAAD.Furthermore, PAAD should also benefit from a third-generation ADC. Indeed, PCA062 is a second-generation ADC, that is, with the anti-CDH3 antibody being directly conjugated to a noncleavable maytansine-derived SMCC-DM1 linker-payload. Such ADCs were designed to kill tumors expressing high levels of target. A noncleavable linker limits the bystander killing effect of adjacent cancer cells and is less suitable in tumors with heterogeneous or moderate target expression, like CDH3 in PAAD. Finally, a DM1 payload might be less optimal than a topoisomerase-I inhibitor, as already shown in pancreatic cancer cells (3). Third-generation ADCs that associate cleavable linkers, which increase the bystander effect, and topoisomerase I inhibitors are expected to be more efficient, as exemplified by the increased efficacy of trastuzumab–deruxtecan versus Trastuzumab–DM1 in HER2+ colon and gastric cancers (4, 5).Thus, we suggest that a third-generation CDH3-ADC might be better suited in solid tumors such as PAAD and deserves further evaluation.M. Lopez is a cofounder and shareholder of Emergence Therapeutics. No disclosures were reported by the other authors.