157P - CLDN18 fusions rather than expression is a biomarker related to the efficacy of paclitaxel in patients with ovarian metastasis of gastric cancer

Background

Paclitaxel inhibits cellular microtubule depolymerisation and has antitumor effects in patients with ovarian metastases from gastric cancer. CLDN18 is aberrantly expressed in gastric cancer and plays an important role in tumor invasion and metastasis. However, the relationship between CLDN18 and the efficacy of paclitaxel is lacking in research.

Methods

A total of 74 GC patients with ovarian metastasis treated with paclitaxel were included in the study. 65 primary gastric and 73 ovarian metastatic tumor samples were collected from the enrolled patients, including 64 pairs of matched primary gastric and metastatic ovarian tumor samples. WES was performed with a mean coverage depth of 187x (range: 108-344x) for tumor samples. Immunohistochemistry (IHC) was performed on tissue samples to detect CLDN18 expression.

Results

Of 74 patients, 41 responded to paclitaxel (the "effective" group), while 33 did not respond (the "ineffective" group). Statistical analysis revealed that mutations in RHOA (p = 0.037), AFF2 (p = 0.028), PIK3CD (p = 0.028), and TAF1L (p = 0.028) were associated with the ineffectiveness of paclitaxel treatment, while mutations in SALL4 (p = 0.036), CCDC105 (p = 0.018), and CLDN18 (p = 0.036) were associated with a positive response to paclitaxel treatment. IHC was performed on six CLDN18 fusion patients and twelve patients with different responses to paclitaxel. Of CLDN18 fusion patients, four were CLDN18-positive and two were CLDN18-negative, indicating that CLDN18 fusion did not correlate with CLDN18 expression. CLDN18 expression was positive in nine patients who responded to paclitaxel treatment and four patients who did not, and negative in two patients who responded and three who did not. Statistical analysis indicated that there was no significant correlation between the effect of paclitaxel treatment and the expression of CLDN18 (9/11 vs. 4/7, p = 0.17).

Conclusions

We have identified potential biomarkers, including CLDN18, that could predict the efficacy of paclitaxel treatment based on the differential response to treatment. We have also confirmed that CLDN18 fusion is a potential predictive biomarker for paclitaxel treatment response, independent of CLDN18 expression.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.