Folate receptor alpha (FRA) expression in breast cancer: identification of a new molecular subtype and association with triple negative disease

IMGN853, a Folate Receptor-α (FRα)-Targeting Antibody-Drug Conjugate, Exhibits Potent Targeted Antitumor Activity against FRα-Expressing Tumors

A majority of ovarian and non-small cell lung adenocarcinoma cancers overexpress folate receptor α (FRα). Here, we report the development of an anti-FRα antibody-drug conjugate (ADC), consisting of a FRα-binding antibody attached to a highly potent maytansinoid that induces cell-cycle arrest and cell death by targeting microtubules. From screening a large panel of anti-FRα monoclonal antibodies, we selected the humanized antibody M9346A as the best antibody for targeted delivery of a maytansinoid payload into FRα-positive cells. We compared M9346A conjugates with various linker/maytansinoid combinations, and found that a conjugate, now denoted as IMGN853, with the N-succinimidyl 4-(2-pyridyldithio)-2-sulfobutanoate (sulfo-SPDB) linker and N(2')-deacetyl-N(2')-(4-mercapto-4-methyl-1-oxopentyl)-maytansine (DM4) exhibited the most potent antitumor activity in several FRα-expressing xenograft tumor models. The level of expression of FRα on the surface of cells was a major determinant in the sensitivity of tumor cells to the cytotoxic effect of the conjugate. Efficacy studies of IMGN853 in xenografts of ovarian cancer and non-small cell lung cancer cell lines and of a patient tumor-derived xenograft model demonstrated that the ADC was highly active against tumors that expressed FRα at levels similar to those found on a large fraction of ovarian and non-small cell lung cancer patient tumors, as assessed by immunohistochemistry. IMGN853 displayed cytotoxic activity against FRα-negative cells situated near FRα-positive cells (bystander cytotoxic activity), indicating its ability to eradicate tumors with heterogeneous expression of FRα. Together, these findings support the clinical development of IMGN853 as a novel targeted therapy for patients with FRα-expressing tumors.

Folate receptor-α (FOLR1) expression and function in triple negative tumors

Folate receptor alpha (FOLR1) has been identified as a potential prognostic and therapeutic target in a number of cancers. A correlation has been shown between intense overexpression of FOLR1 in breast tumors and poor prognosis, yet there is limited examination of the distribution of FOLR1 across clinically relevant breast cancer subtypes. To explore this further, we used RNA-seq data from multiple patient cohorts to analyze the distribution of FOLR1 mRNA across breast cancer subtypes comprised of estrogen receptor positive (ER+), human epidermal growth factor receptor positive (HER2+), and triple negative (TNBC) tumors. FOLR1 expression varied within breast tumor subtypes; triple negative/basal tumors were significantly associated with increased expression of FOLR1 mRNA, compared to ER+ and HER2+ tumors. However, subsets of high level FOLR1 expressing tumors were observed in all clinical subtypes. These observations were supported by immunohistochemical analysis of tissue microarrays, with the largest number of 3+ positive tumors and highest H-scores of any subtype represented by triple negatives, and lowest by ER+ tumors. FOLR1 expression did not correlate to common clinicopathological parameters such as tumor stage and nodal status. To delineate the importance of FOLR1 overexpression in triple negative cancers, RNA-interference was used to deplete FOLR1 in overexpressing triple negative cell breast lines. Loss of FOLR1 resulted in growth inhibition, whereas FOLR1 overexpression promoted folate uptake and growth advantage in low folate conditions. Taken together, our data suggests patients with triple negative cancers expressing high FOLR1 expression represent an important population of patients that may benefit from targeted anti-FOLR1 therapy. This may prove particularly helpful for a large number of patients who would typically be classified as triple negative and who to this point have been left without any targeted treatment options.

Folate receptor α associated with triple-negative breast cancer and poor prognosis

CONTEXT

Folate receptor α (FRA) has been shown to be selectively expressed in several types of human cancer, including breast cancer. Currently, several FRA target therapies are under intensive study.

OBJECTIVE

To investigate the expression pattern of FRA in a large cohort of patients with breast cancer and analyze its relationship with different clinicopathologic features, with expression of several key biomarkers, and with clinical outcome.

DESIGN

Four hundred forty-seven cases of infiltrating ductal carcinoma diagnosed between 1997 and 2008 at the University of Rochester Medical Center were identified and reviewed, and 25 blocks of tissue microassays were constructed. The association between expression of FRA and clinicopathologic features; expression of estrogen receptor (ER), progesterone receptor (PR), HER2/neu, and Ki-67; and clinical outcome of these tumors were evaluated.

RESULTS

The expression of FRA was significantly associated with tumors with high histologic grade, higher nodal stages, ER/PR negativity, and high proliferative activity (Ki-67 ≥ 15%), and was independent of HER2/neu overexpression. In all, 74% of ER/PR-negative and 80% of triple-negative breast cancers expressed FRA. The expression of FRA was significantly associated with a worse disease-free survival.

CONCLUSIONS

Our data demonstrate that a significant subgroup of ER/PR-negative and triple-negative breast cancers express FRA, and its expression is associated with worse clinical outcome.

Serum folate receptor alpha, mesothelin and megakaryocyte potentiating factor in ovarian cancer: association to disease stage and grade and comparison to CA125 and HE4

Background

Evaluate and compare the utility of serum folate receptor alpha (FRA) and megakaryocyte potentiating factor (MPF) determinations relative to serum CA125, mesothelin (MSLN) and HE4 for the diagnosis of epithelial ovarian cancer (EOC).

Methods

Electrochemiluminescent assays were developed for FRA, MSLN and MPF and used to assess the levels of these biomarkers in 258 serum samples from ovarian cancer patients. Commercial assays for CA125 and HE4 were run on a subset of 176 of these samples representing the serous histology. Data was analyzed by histotype, stage and grade of disease. A comparison of the levels of the FRA, MSLN and MPF biomarkers in serum, plasma and urine was also performed in a subset of 57 patients.

Results

Serum and plasma levels of FRA, MSLN and MPF were shown to be highly correlated between the two matrices. Correlations between all pairs of markers in 318 serum samples were calculated and demonstrated the highest correlation between HE4 and MPF, and the lowest between FRA and MPF. Serum levels of all markers showed a dependence on both stage and grade of disease. A multi-marker logistic regression model was developed resulting in an AUC=0.91 for diagnosis of serous ovarian cancer, a significant improvement over the AUC for any of the individual markers, including CA125 (AUC=0.84).

Conclusions

FRA has significant potential as a biomarker for ovarian cancer, both as a stand-alone marker and in combination with other known markers for EOC. The lack of correlation between the various markers analyzed in the present study suggests that a panel of markers can aid in the detection and/or monitoring of this disease.

Interobserver Agreement and Assay Reproducibility of Folate Receptor α Expression in Lung Adenocarcinoma: A Prognostic Marker and Potential Therapeutic Target

Context.—Lung cancer is the leading cause of cancer deaths in the United States and globally. Folate-targeted drugs are among the promising new targeted therapies for lung cancer, provided predictive biomarkers can be identified for optimal patient selection.

Objective.—To evaluate the interobserver agreement and reproducibility of an immunohistochemistry assay for folate receptor α as a potential predictive marker for folate-targeted therapies.

Design.—Immunohistochemistry using anti–folate receptor α antibody 26B3 was performed on formalin-fixed, paraffin-embedded tissues. The M-score, a semiquantitative measure of staining intensity and proportion of tumor cells staining, was determined for each specimen. Interobserver agreement was assessed using lung adenocarcinoma specimens stained at a single site and evaluated by 3 independent pathologists. Interinstrument reproducibility assessed 20 specimens stained by 3 different automated stainers. Interlaboratory agreement was determined on 5 specimens, repeatedly stained on each of 5 days, at 3 different study sites.

Results.—Folate receptor α expression was identified in 39 of 54 cases of lung adenocarcinoma (72%) and 4 of 37 cases of lung squamous cell carcinoma (11%). Agreement among 3 pathologists was found in 24 of 26 cases (92%). Interinstrument reproducibility was observed in 19 of 20 cases (95%). Agreement among 3 laboratories was found for 49 of 50 specimens (98%).

Conclusions.—Immunostaining of folate receptor α in lung adenocarcinomas is reproducible across staining platforms and among laboratories. Agreement among pathologists is achieved using a semiquantitative scoring method. An accurate and convenient method for determining folate receptor α expression offers a potentially invaluable tool for selecting patients for folate-targeted therapies.

Folate receptor alpha (FRA) expression remains unchanged in epithelial ovarian and endometrial cancer after chemotherapy

OBJECTIVE

Based on its expression profile, folate receptor alpha (FRA) is an attractive candidate for targeted diagnostics and therapeutics. However, applicability of these agents in residual or recurrent disease could be influenced by chemotherapy. We evaluated whether chemotherapy modified FRA expression in non-mucinous epithelial ovarian (EOC) and endometrial carcinoma (EC).

METHODS

FRA staining was evaluated by immunohistochemistry, using MAb 26B3, in 81 patients (41 EOCs and 40 ECs) and 17 control tissues (5 benign ovarian cysts, 5 normal ovarian, and 7 normal endometrial tissues). Chemotherapy effect was evaluated in 42 patients (30 paired samples at primary and interval debulking surgery and 12 from primary and recurrent disease). FRA expression was assessed using a semi-quantitative staining algorithm, the M-score (range 0-50).

RESULTS

Median difference in M-score between tumor and control samples was 27.5 for EOC (95% CI 10.0 to 45.0) and 6.7 for EC (95% CI -6.7 to 21.7). Paired samples from both primary and interval debulking surgery did not differ in FRA expression in EOC (median difference of M-score between paired samples of 0.0 [95% CI -2.6 to 2.6]). Recurrent EOC tumors reflected FRA status at diagnosis (median difference of M-score between paired samples of 3.3 [95% CI -7.0 to 13.6]).

CONCLUSIONS

This study shows no significant difference in FRA expression after chemotherapy, strengthening the rationale for FRA targeted diagnostics and therapeutics in FRA expressing tumors, whether newly diagnosed or at recurrence.