Claudin18.2 as a potential therapeutic target for primary ovarian mucinous carcinomas and metastatic ovarian mucinous carcinomas from upper gastrointestinal primary tumours

Ovarian mucinous carcinomas, including primary and metastatic cancers, account for 10 ~ 15% of epithelial ovarian cancers [1, 2], but primary ovarian mucinous carcinoma (POMC) represents only approximately 1 ~ 3% of epithelial ovarian tumours [1, 3]. Metastatic ovarian tumours may be derived from tumours in all parts of the body, and the most common primary sites are the gastrointestinal tract, breast, lung, and cervix, with gastrointestinal primary tumours, including upper and lower gastrointestinal tract tumours, being the most common origin [4, 5].

The 5-year survival rate of stage I POMC is 92%; however, if metastasis to the abdominal cavity occurs, in stage III and IV, the overall survival time is only 12–33 months, which is significantly lower than that of advanced-stage high-grade serous carcinoma (35 ~ 60 months) [3]. The prognosis of metastatic mucinous ovarian carcinoma is very poor, as patients usually die within 2 years of diagnosis; prognosis seems to be associated with the origin of ovarian metastatic carcinoma and optimal metastasectomy [6, 7]. The 5-year survival rates of metastasectomy-treated ovarian mucinous carcinoma originating from gastric, colorectal, and breast tumours are 0, 20.7 and 22.2%, respectively [7].

Since the morphological features of POMC and metastatic mucinous ovarian carcinoma are similar, differential diagnosis is difficult [8]. Furthermore, some clinicians suggest that oxaliplatin and 5-fluorouracil regimens are appropriate for postoperative treatment because primary mucinous carcinomas of the ovary are similar to gastrointestinal tumours [9]. The National Comprehensive Cancer Network (NCCN) guidelines recommend that POMC patients be treated with gastrointestinal malignancy chemotherapy regimens such as 5-fluorouracil and oxaliplatin, but their effectiveness has not been assessed in a large-scale study. Therefore, effective therapeutic approaches for primary and metastatic ovarian mucinous carcinoma deserve further study.

Claudin18.2 is highly expressed in gastric and oesophageal adenocarcinoma [10‐12]. Claudin18.2 is a highly promising therapeutic target for claudin18.2-overexpressing tumours. Zolbetuximab (claudiximab), the first developed drug to target claudin18.2, is a chimeric IgG1 monoclonal antibody that specifically binds to claudin18.2 on the surface of tumour cells, triggering antibody-dependent cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC), apoptosis and inhibition of cell proliferation. Preclinical studies successfully demonstrated that it has a powerful ability to clear cancer cells and control disease. Subsequently, its clinical efficacy and safety were evaluated by phase I/II/III trials [13‐15]. Because the high specificity of claudin18.2 helps T cells recognize tumours, it is used for chimeric antigen receptor T-cell therapy (CAR-T). Patient-derived xenograft (PDX) models have been used to validate efficacy in vivo and thus may lead to breakthroughs, demonstrating claudin18.2-specific CAR T cells as a promising therapeutic strategy for claudin18.2-positive tumours, especially gastric cancer [16‐18].

Claudin18 expression has been studied in borderline ovarian mucinous tumours and benign ovarian mucinous tumours [19, 20], but there is no study on claudin18.2 expression in ovarian mucinous carcinomas. Therefore, in our study, we investigated claudin18.2 expression in both POMCs and MGMCs to explore its potential role in targeted therapy.

The SEER data from 1973–2015 showed that mucinous epithelial ovarian cancer patients exhibited better outcomes in the localized stage (FIGO I-A, I-B, I-not otherwise specified [NOS]) and regional stage (FIGO I-C, II-A, II-B, II-C, II-NOS) than in the distant stage (FIGO III and IV); for patients diagnosed at distant disease, the 3-, 5- and 10-year survival rates for mucinous epithelial ovarian cancer were 22.1, 16.1 and 10.9%, which were even worse than those for high-grade serous ovarian carcinoma (51.7, 31.9 and 15.5%) [21], which may be related to its poor response to chemotherapy and indolent biological behaviour [3, 22‐24]. Improving the resection rate of advanced POMC can help to improve therapeutic efficacy [3]. Moreover, new strategies for the management of high-risk localized disease and advanced-stage POMC need to be assessed in clinical trials.

The prognosis of patients with gastric tumour-derived ovarian carcinomas was significantly worse than that of patients with colorectal tumour-derived ovarian carcinomas, and the survival time was 11 vs. 21.5 months [25]. Kajiyama et al. reported that 91.0% of patients with gastric tumour-derived ovarian carcinomas had recurrence, and 81.2% died of the disease during follow-up. However, 76.3% of patients with nongastric carcinoma developed recurrence, and 64.4% died of the disease. The median overall survival (OS) times were 8.1 and 28.1 months in patients with gastric cancer and nongastric cancer, respectively [26]. Among metastatic mucinous ovarian carcinoma types, those derived from the upper gastrointestinal tract have the worst prognosis, and effective therapeutics for these patients are worth exploring.

Claudin18 has two isoforms, claudin18.1 and claudin18.2, specific for pulmonary and gastric tissue, respectively. Claudin18.2 overexpression has been identified in several other types of cancers, including gastric, esophageal pancreatic adenocarcinomas [27‐30]. In this study, we found a higher rate of claudin18.2 expression in POMCs and in MGMCs of upper gastrointestinal tract origin than in MGMCs of lower gastrointestinal tract origin. The positive expression rates of claudin18.2 in POMCs and borderline mucinous tumours were 86.4% and 80.0%, respectively. Positive expression was observed in 91.7% of advanced POMCs (FIGO III-IV). However, expression was negative in all ovarian serous tumours. In MGMCs, the positive expression rate of metastatic ovarian carcinomas from the upper gastrointestinal tract was 69.6%. Positive expression was observed included 15 cases of metastases from gastric cancer (15/22,68.2%) and 1 case of metastasis from pancreatic cancer (1/1), while no claudin18.2 expression was found in MGMC cases derived from the lower gastrointestinal tract.

Normally, claudin18.2 is expressed only on the surface of epithelial cells differentiated from the gastric mucosa, and claudin 18.2 expression in normal mucosal cells is restricted to the interior of the tight junction complex. However, after the occurrence of malignancy, the tight junction protein is destroyed, exposing the claudin18.2 epitope on the surface of the tumour cells, and making it an extremely ideal drug target.

Zolbetuximab(IMAB362) was the first mAb to target claudin18.2. In the FAST study, first-line treatment with zolbetuximab in gastric cancer led to a 50% increase in the progression-free survival (PFS) and a twofold increase in overall survival (OS) time [31]. Claudin18.2-positive gastric/gastroesophageal junction/oesophageal adenocarcinoma patients received first-line EOX (epirubicin plus oxaliplatin plus capecitabine) alone or combined with zolbetuximab. For patients with claudin18.2 expression in ≥ 70% of tumour cells, the median PFS with the zolbetuximab + EOX regimen was 9.0 months, and the median OS was 16.5 months. This group of patients benefitted most from zolbetuximab treatment. On the other hand, for patients with 40 ~ 69% claudin18.2 positive tumour cells, the median PFS with the EOX regimen was 5.7 months, and the median OS was 8.9 months [31]. In terms of efficacy, the higher the proportion of cancer cells expressing claudin18.2 is the more significant the advantage of the zolbetuximab regimen. Moreover, in the MONO study, zolbetuximab was used as monotherapy in patients with metastatic or progressive gastric/gastroesophageal junction/oesophageal cancer, including patients with moderate or strong claudin18.2 membrane staining in 50% of tumour cells. The regimen showed a 9% overall response rate (ORR) and a 23% clinical benefit rate (partial response and stable disease). In the subgroup of patients expressing claudin18.2 in ≥ 70% of the tumour cells, the ORR was increased to 14% [14]. Thus, the inclusion criterion for positive claudin18.2 expression in cancer cells was increased to 75% in many recent clinical studies [32]. In our study, a positive staining rate greater than 75% with a staining intensity score greater than 2 + was also applied as a critical value for assessing claudin18.2 expression. In the POMC group, 72.7% of patients had more than 75% claudin18.2-positive cells with moderate or strong staining, and 75.0% (9/12) of patients had in advanced-stage (stage III/IV) POMCs. For MGMCs of upper gastrointestinal tract origin, among 22 patients with tumours derived from gastric cancer, 6 patients (27.3%) had ovarian metastatic tissue with more than 75% stained tumour cells with intensity score ≥ 2 + . Moreover, one patient with pancreatic metastatic mucinous ovarian carcinoma had a score of 3 + in almost all tumour cells.

There are still limitations to this study. For both POMCs and MGMCs derived from upper gastrointestinal tract primary tumours, we should further compare claudin18.2 expression levels between primary and metastatic tumours. The combined detection of claudin18.2 and other targeted therapeutic targets, such as HER-2, PD-L1, MMR, and tumour-infiltrating lymphocytes (TILs), will facilitate the exploration of combined multitarget treatment strategies and the prediction of the response to these therapies for patients with POMCs and MGMCs.

According to the clinicaltrials.gov and chinadrugtrials.gov websites, dozens of clinical studies are currently ongoing to apply claudin18.2-targeted therapeutic strategies for solid tumours with high claudin18.2 expression. These claudin18.2-targeting agents include monoclonal antibodies (mAbs), bispecific antibodies (BsAbs), antibody-conjugated drugs (ADCs), and CAR-T immunotherapy. It is hoped that further clinical studies will confirm that patients with POMCs and MGMCs from the upper gastrointestinal tract with high claudin18.2 expression can benefit from therapeutic strategies targeting claudin18.2.