Characterizing therapeutic target antigen expression in anaplastic carcinoma of the ovary

Highlights

• •We evaluated HER2, FOLR1, and TROP2 expression in anaplastic carcinoma of the ovary.
• •Expression of these antigens was low or negative in all 9 patients in our cohort.
• •There is an urgent need for novel treatment approaches in this rare cancer.

Abstract

Objective

Novel treatment strategies are needed for patients with anaplastic carcinoma of the ovary given the high incidence of chemoresistance to conventional systemic therapies. Here we sought to evaluate HER2, FOLR1, TROP2, and mismatch repair protein expression as well as the tumor mutational burden of anaplastic carcinoma of the ovary.

Methods

We retrospectively identified a total of 9 anaplastic ovarian carcinomas in institutional databases from 2013 to 2023, including one case from a collaborating institution. Cases with adequate tissue available underwent immunohistochemical analysis for HER2, FOLR1, TROP2, and mismatch repair proteins MLH1, MSH2, MSH6, and PMS2. Tumor mutational burden was obtained from tumor-normal panel sequencing.

Results

Anaplastic carcinoma of the ovary demonstrated low or absent expression of FOLR1, HER2, and TROP2 protein. FOLR1 expression in the two cases was weak and below currently used clinical cutoffs for mirvetuximab soravtansine eligibility. HER2 expression was low in the two anaplastic carcinomas (1–2+). TROP2 expression was low in one case (H-score of 60) and negative in the other five. All cases were mismatch repair proficient by immunohistochemistry and had low tumor mutational burden (median, 3.3 mut/Mb; range, 0.8–6.9).

Conclusion

Only a subset of anaplastic carcinomas of the ovary express targetable tumor antigens at low levels, suggesting that these may have limited benefit from antibody–drug conjugates. Likewise, due to mismatch repair proficiency and low tumor mutational burden, immunotherapy is unlikely to be effective in this rare disease.

1. Introduction

Anaplastic carcinoma of the ovary is a rare, aggressive cancer that usually presents as a sarcomatous mural nodule within a mucinous ovarian carcinoma (Sullivan et al., 2025). The literature regarding this rare tumor type consists of case reports and series upon which our previously published work expanded. The clinical outcomes in our patient series showed limited efficacy of standard treatment options, where five of six patients with progression or recurrence had platinum-refractory disease and the sixth had a progression-free interval of 6.8 months (Sullivan et al., 2025). We observed a high prevalence of KRAS (80%), TP53 (60%), and CDKN2A (60%) alterations, similar to those seen in mucinous ovarian carcinomas, consistent with similar findings reported by researchers from the Brigham and Women’s Hospital (Sullivan et al., 2025, Chapel et al., 2021).

Antibody–drug conjugates represent an exciting new therapeutic option for many patients with gynecologic cancers (Bujnak et al., 2025). The U.S. Food and Drug Administration approval of mirvetuximab soravtansine and fam-trastuzumab deruxtecan in recurrent gynecologic malignancies with high protein expression of FOLR1 and HER2, respectively, has paved the way for exploration of other therapeutic targets in these cancers (Bujnak et al., 2025). The efficacy of fam-trastuzumab deruxtecan in multiple solid tumors was demonstrated in the phase 2 DESTINY-PanTumor02 trial and was found to correlate with HER2 protein expression as defined by immunohistochemistry (Meric-Bernstam et al., 2024, Oaknin et al., 2024). Similarly, mirvetuximab soravtansine was found to have an overall survival benefit in recurrent platinum-resistant, FOLR1-positive high-grade serous ovarian cancer in a phase 3 trial published in 2023, which led to its U.S. Food and Drug Administration approval for this indication (Moore et al., 2023). Given the promising results seen in the more common gynecologic malignancies, researchers have sought to determine if patients with rare subtypes may have comparable levels of target antigen expression and could potentially benefit from these novel therapies. For example, research from our group showed that 40% of tumors from patients with low-grade serous ovarian cancer exhibited high FOLR1 expression, a rate comparable to that observed in high-grade serous ovarian cancer (Manning-Geist et al., 2024). TROP2 is another therapeutic target of interest in several disease sites and has previously been shown to be expressed in gynecologic malignancies (Bignotti et al., 2011, Moufarrij et al., 2025, Lopez et al., 2020). Sacituzumab govitecan is an antibody–drug conjugate that targets TROP2 and has been shown in preclinical models to have activity in gynecologic malignancies, supporting our rationale for subjecting these tumors to TROP2 immunohistochemistry (Moufarrij et al., 2025, Lopez et al., 2020).

In this work, we sought to determine the potential applicability of currently approved antibody–drug conjugates and immunotherapy in the treatment of anaplastic carcinoma of the ovary. We conducted immunohistochemistry analyses to assess the expression of HER2, FOLR1, and TROP2 in these malignancies and evaluated the status of mismatch repair proteins MLH1, MSH2, MSH6, and PMS2 in an effort to generate hypotheses for novel treatment approaches in this rare cancer. MSK-IMPACT data were used to determine tumor mutational burdens.

2. Methods

This study was approved by the Memorial Sloan Kettering Cancer Center Institutional Review Board. Anaplastic carcinomas of the ovary were retrospectively collected from our institutional databases between 2013 and 2023. Patients were also enrolled from a collaborating institution (Walter Reed National Military Medical Center). For cases with available material, representative slides were subjected to immunohistochemistry analysis of FOLR1, HER2, TROP2, and mismatch repair proteins (MLH1, MSH2, MSH6, and PMS2) as previously described (Johannet et al., 2025). Priority was given to FOLR1 and HER2 immunohistochemistry, with TROP2 and mismatch repair evaluation considered a secondary objective. All patients provided written consent for clinical and tissue research. Tumor mutational burden was obtained from tumor-normal sequencing data (MSK-IMPACT) (Sullivan et al., 2025).

FOLR1 expression was considered positive if ≥25% of cells had ≥2+ staining (Gilbert et al., 2023). HER2 expression was graded using the American Society of Clinical Oncology/College of American Pathologists gastroesophageal HER2 guidelines on a 0–3+ scale (Bartley et al., 2017). TROP2 expression was reported using the H-score method ([1 × % tumor cells with weak staining] + [2 × % tumor cells with moderate staining] + [3 × % tumor cells with strong staining]) (Moufarrij et al., 2025). An expert pathologist (M.H.C.) reviewed all immunohistochemistry analyses. We employed descriptive statistics.

3. Results

Tissue samples from nine patients with anaplastic carcinoma of the ovary were available for analysis. First, we assessed expression of potential antibody–drug conjugate targets, FOLR1 and HER2. All tumors had minimal FOLR1 expression, with seven of the nine showing zero staining and two showing weak staining in a small percentage of tumor cells (1% and 10%, respectively) (Fig. 1). Similarly, seven anaplastic carcinomas of the ovary (78%) had a HER2 immunohistochemistry score of 0, and two (22%) had 1+ and 1–2+ expression (Table 1).

Fig. 1.

Immunohistochemical analysis for antibody–drug conjugate targets in ovarian anaplastic carcinoma. (A-D) Representative photomicrographs from Case AP7, exhibiting (A) poorly differentiated, sarcomatoid morphology (hematoxylin and eosin), (B) weak FOLR1 expression in ∼10% of tumor cells, (C) HER2 immunohistochemistry score of 2+, and (D) absence of TROP2 expression. (E) In anaplastic ovarian carcinomas, TROP2 often stains the co-existing mucinous borderline tumor component but not the anaplastic carcinoma component (Case AP9). However, in Case AP4 we observed (F) patchy, moderate-intensity TROP2 expression within the anaplastic carcinoma (H-score = 60).

Table 1.

Target antigen expression in anaplastic ovarian carcinoma.

Study ID	Age at diagnosis (years)	Stage	HER2	FOLR1	TROP2	MMR	TMB (mut/Mb)
1	19	IC2	0	0	0	Proficient	4.1
2	31	IIIA	0	0	NP	Proficient	0.8
3	31	IIIA1	0	0	60	Proficient	2.6
4	31	IV	0	0	0	Proficient	3.3
5	53	II	0	0	0	Proficient	2.5
6	47	IIIC	0	0	0	Proficient	6.9
7	21	IA	1–2+	10% weak	0	Proficient	1.8
8	60	IIA	1+	0	NP	NP	4.9
9	34	IA	0	1%	NP	NP	3.9

MMR, mismatch repair; NP, not performed; TMB, tumor mutational burden.

Next, we assessed for TROP2 expression, mismatch repair deficiency, and tumor mutational burden. We evaluated TROP2 expression in six cases with available material. One of these cases had low TROP2 expression (H-score = 60) and the other five were negative (H-score = 0). We performed mismatch repair immunohistochemistry testing in seven of the nine patients, and all seven tumors had proficient/retained mismatch repair protein expression of all four proteins (MLH1, MSH2, MSH6, and PMS2). Consistent with these findings, we discovered that all cases were microsatellite stable via MSK-IMPACT. Tumor mutational burden data were available for all cases, revealing that anaplastic carcinoma of the ovary had low tumor mutational burden, with a median of 3.3 mut/Mb (range, 0.8–6.9).

4. Discussion

Anaplastic carcinoma of the ovary demonstrates resistance to traditional chemotherapeutic treatments, as evidenced by the high rate of platinum-refractory disease observed in our previous study of patients with advanced-stage disease (Sullivan et al., 2025). This underscores the urgent need for novel therapeutic strategies to effectively treat patients with this aggressive and rare form of ovarian cancer. Given our previous findings of high rates of KRAS, TP53, and CDKN2A tumor genomic alterations, we proposed consideration of novel KRAS-targeted therapies as potential options for this patient population. In the current work, we sought to generate further hypotheses by assessing the potential utility of currently approved ovarian cancer antibody–drug conjugates (mirvetuximab soravtansine and fam-trastuzumab deruxtecan) in this patient population. Secondarily, due to the molecular similarities between these cancers and their mucinous/gastrointestinal counterparts, we performed hypothesis-generating immunohistochemistry staining for mismatch repair proteins to investigate the possible role of immunotherapy. We also evaluated TROP2 expression in cases with sufficient material.

We found that all patients’ tumors (n = 9) had low or absent expression of FOLR1, with only two showing weak staining. Similarly, only two (22%, n = 9) patients showed any expression of HER2, both of which were HER2-low. TROP2 expression was also negative or low in all cases evaluated. Although disappointing, these findings further emphasize the urgent need for novel treatment approaches in this rare cancer, as even the most promising molecular targets identified in other epithelial ovarian cancers are not highly expressed in these tumors. Furthermore, we found no cases exhibiting mismatch repair deficiency, microsatellite instability, or high tumor mutational burden, indicating that immune modulation may likewise be ineffective in this group.

Emerging literature on antibody–drug conjugates indicates that clinical benefit can still be achieved in malignancies with low levels of target antigen expression. One proposed mechanism underlying this phenomenon is a bystander effect, whereby payload release from antigen-positive cells results in cytotoxicity of neighboring antigen-negative cells. Additional contributing factors may include tumor heterogeneity, dynamic antigen expression, and the high potency of the cytotoxic payload. Mirvetuximab soravtansine was initially approved for malignancies with ≥75% of cells with ≥2+ staining; however, based on subsequent studies, it has since been listed by the National Comprehensive Cancer Network for tumors with ≥25% of cells staining positive for FOLR1 in combination with bevacizumab (Moore et al., 2023, Gilbert et al., 2023). Similarly, the recent phase 2 STATICE trial demonstrated the efficacy of fam-trastuzumab deruxtecan in endometrial carcinosarcoma across varying levels of HER2 immunohistochemistry expression, with a 70% overall response rate in HER2-low (immunohistochemistry score 1+) populations (Nishikawa et al., 2023). Although data showing benefit in anaplastic ovarian cancer remain lacking, similar patterns observed in other gynecologic malignanacies suggest that antibody–drug conjugates may retain activity even in the setting of low or hetergenous target antigen expression.

Strengths of this study include being the first to report on these tumor-associated antigen antibody–drug conjugate targets in this rare histologic subtype of ovarian cancer. Given the rare nature of this disease, the small sample size is an inherent limitation; however, these data remain hypothesis-generating and contribute meaningful insight in an area with limited existing literature. Where feasible, clinical tumor-associated antigen testing should be considered, as occasional clinical benefit may be possible despite low target antigen expression rates, while simultaneously allowing further retrospective study in this rare disease.

5. Conclusions

Most anaplastic ovarian carcinomas do not express FOLR1, HER2, or TROP2, and most are mismatch repair proficient with low tumor mutational burden. Although there may be some activity in this rare tumor subtype with currently approved antibody–drug conjugates and immunotherapy, the benefit is likely to be modest.

CRediT authorship contribution statement

Mackenzie W. Sullivan: Writing – review & editing, Writing – original draft, Formal analysis, Data curation, Conceptualization. M. Herman Chui: Writing – review & editing, Data curation. Alexa N. Kanbergs: Writing – review & editing. Hunter Green: Writing – review & editing, Data curation. Kaitlyn Gill: Writing – review & editing, Data curation. Pier Selenica: Writing – review & editing, Data curation. Achim A. Jungbluth: Writing – review & editing. Kara C. Long: Writing – review & editing. Yukio Sonoda: Writing – review & editing. Jennifer J. Mueller: Writing – review & editing. Christopher Tarney: Writing – review & editing. Rachel N. Grisham: Writing – review & editing. Vicky Makker: Writing – review & editing. Nadeem R. Abu-Rustum: Writing – review & editing. Britta Weigelt: Writing – review & editing, Formal analysis, Conceptualization. Roisín E. O’Cearbhaill: Writing - review & editing, Formal analysis, Conceptualization.

Funding

Research reported in this publication was supported in part by a National Institutes of Health (NIH)/National Cancer Institute (NCI) Cancer Center Support Grant (P30 CA008748; Biobank and Pathology Core). BW is supported in part by Cycle for Survival, NIH/NCI (P50 CA247749), and Breast Cancer Research Foundation grants.

Consent statement

This retrospective study was approved by the Institutional Review Board at Memorial Sloan Kettering Cancer Center, and written informed consent was obtained from all participants.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: A.A. Jungbluth reports royalties from Memorial Sloan Kettering Cancer Center, Ludwig Institute for Cancer Research; R.N. Grisham reports honoraria from GlaxoSmithKline, AstraZeneca, Natera, Springworks, Corcept, MJH, and PER. V. Makker reports grants from Merck, Eisai, AstraZeneca, Clovis, Bayer, Takeda, Duality, Zymeworks, Karyopharm, Faeth, Bristol-Myers Squibb, Lilly, Cullinan paid to the institution; travel support from Eisai, Merck, AstraZeneca; unpaid consulting/advisory role with Duality, Novartis, Morphosys, AstraZeneca, ArQule, Eisai, Cullinan, Clovis, Karyopharm, GSK, Merck, Faeth, Jazz, Immunocore, Iteos Therapeutics, Ideaya, Kartos Therapeutics, Lilly, Moreo, Prelude, Takeda, Zymeworks; and other financial or nonfinancial interests in IBM. N.R. Abu-Rustum reports research funding from GRAIL paid to the institution. B. Weigelt reports research support by Repare Therapeutics and SAGA Diagnostics, and employment of a direct family member at AstraZeneca. R.E. O’Cearbhaill reports institutional research grants from Arsenal Bio, AstraZeneca/Merck, Atara Biotherapeutics/Bayer, Context Therapeutics, Fate Therapeutics, Genentech, Genmab, GSK, Gynecologic Oncology Group Foundation, Juno Therapeutics, Kite/Gilead, Ludwig Institute for Cancer Research, Lyell Therapeutics, Marker Therapeutics, Mural Oncology, OnCusp Therapeutics, Regeneron, Sellas Life Sciences, Stemcentrx, Solve Therapeutics, Syndax, TapImmune, and TCR2 Therapeutics; participating in advisory boards with Bayer, Carina Biotech, Corcept Therapeutics, Immunogen/Abbvie, Miltenyi, Loxo, Pfizer, Regeneron, R-Pharm, Seattle Genetics, and Tesaro/GSK; personal fees from GOG Foundation; travel fees from Hitech Health. The other authors do not have potential conflicts of interest to declare.