Efficacy and Safety of Anti-Trop-2 Antibody Drug Conjugate Sacituzumab Govitecan (IMMU-132) in Heavily Pretreated Patients With Metastatic Triple-Negative Breast Cancer

Abstract

Purpose Trop-2, expressed in most triple-negative breast cancers (TNBCs), may be a potential target for antibody-drug conjugates. Sacituzumab govitecan, an antibody-drug conjugate, targets Trop-2 for the selective delivery of SN-38, the active metabolite of irinotecan. Patients and Methods We evaluated sacituzumab govitecan in a single-arm, multicenter trial in patients with relapsed/refractory metastatic TNBC who received a 10 mg/kg starting dose on days 1 and 8 of 21-day repeated cycles. The primary end points were safety and objective response rate; secondary end points were progression-free survival and overall survival. Results In 69 patients who received a median of five prior therapies (range, one to 12) since diagnosis, the confirmed objective response rate was 30% (partial response, n = 19; complete response, n = 2), the median response duration was 8.9 (95% CI, 6.1 to 11.3) months, and the clinical benefit rate (complete response + partial response + stable disease ≥ 6 months) was 46%. These responses occurred early, with a median onset of 1.9 months. Median progression-free survival was 6.0 (95% CI, 5.0 to 7.3) months, and median overall survival was 16.6 (95% CI, 11.1 to 20.6) months. Grade ≥ 3 adverse events included neutropenia (39%), leukopenia (16%), anemia (14%), and diarrhea (13%); the incidence of febrile neutropenia was 7%. The majority of archival tumor specimens (88%) were moderately to strongly positive for Trop-2 by immunohistochemistry. No neutralizing antibodies to the ADC or antibody were detected, despite repeated cycles developed. Conclusion Sacituzumab govitecan was well tolerated and induced early and durable responses in heavily pretreated patients with metastatic TNBC. As a therapeutic target and predictive biomarker, Trop-2 warrants further research.

Figures

Fig 1.

Efficacy assessments. (A) A waterfall plot that shows the percentage of change in sum of target lesions at the time of best response for 66 of the 69 patients who had at least one postbaseline computed tomography assessment. Forty-eight patients had reductions from baseline, including 23 with at least a 30% reduction (two confirmed complete responses [CRs], 19 confirmed partial responses [PRs], two unconfirmed PRs classified as stable disease by Response Evaluation Criteria in Solid Tumors [RECIST] version 1.1). (B) A swimmer graph that depicts the duration of treatment, onset, and duration of response for 21 responders. Fourteen patients progressed and discontinued treatment at that time (bars without arrows). Six have continuing response and treatment at this time (bars with arrows). One who progressed with new brain lesions after 12.2 months of treatment (star) but after localized radiation to the brain currently continues treatment cycles (bar with arrow) with additional shrinkage to body target lesions. (C) Percentage of change in sum of target lesions during treatment of each of the 21 responders; the gray arrow indicates the time course for the patient with brain metastases. (D) Kaplan-Meier progression-free survival graph for all 69 patients, with 63 having progressed. (E) Kaplan-Meier overall survival graph for all 69 patients, with 33 deceased. PD, progressive disease.

Fig 2.

Examples of patients with objective responses. Case study 1: A 48-year-old woman with an initial diagnosis of triple-negative breast cancer in November 2007 received four prior lines of treatment (two lines in a metastatic setting, including an anti-PD-L1 immune checkpoint inhibitor) and presented with lung and lymph node metastases at the time of enrollment in February 2015. She achieved a partial response that started 1.7 months after initiation of treatment with sacituzumab govitecan, with a best response of 54% reduction at 9.0 months and progression occurring at 14.4 months (partial response duration, 12.7 months). (A) Baseline image of two of the three target lesions: a 24 × 19-mm left-upper-lung mass (arrow) and a mediastinal lymph node (17 × 29-mm; circle). (B) After 16 doses (July 2015), these two target lesions decreased to 13 × 7 and 9 × 19 mm, respectively. (C) Trop-2 expression in an archived tumor specimen by immunohistochemistry that shows 1+ to 2+ staining (overall, 2+). Case study 2: A 67-year-old woman with an initial diagnosis of estrogen receptor–positive breast cancer in 2007 that was treated by lumpectomy and local radiation. In 2012, a local recurrence was treated with neoadjuvant doxorubicin and cyclophosphamide followed by taxol and a mastectomy. One year later, the patient received a diagnosis of metastatic disease, which was biopsy-proven triple-negative breast cancer. She was enrolled in a clinical trial and had stable disease for 3 months, but she discontinued treatment because of progression, including new brain metastases. After stereotactic radiosurgery and three cycles of eribulin, the patient progressed and was enrolled in the current study. Her brain metastases had been stable for > 3 months at the time of enrollment. (D) and (F) Baseline (September 2014) and (E) and (G) 7.2-month follow-up (April 2015) computed tomography scans performed after 22 treatments that showed two target lesions: (D) and (E) axillary and (F) and (G) mediastinal lymph nodes (arrows). Her first computed tomography response assessment in November 2014 showed a 41% reduction (partial response) after just six doses, which was confirmed 1 month later, and at this time, magnetic resonance imaging showed no evidence of intracranial metastases. (H) Photograph of the patient’s skin involvement at the onset of treatment (September 2014) and (I) a photograph that shows a response in December 2014. The patient continued treatment, with 66% reduction as the overall best response recorded in September 2015 and with four of the five nontarget lesions remaining stable and one completely resolved (supraclavicular lymph node). However, in October 2015, two of the nontarget lesions progressed, and she was withdrawn from the study. (J) Trop-2 expression in archived tumor by immunohistology that shows 3+ staining.

Fig A1.

Example of scoring for Trop-2 staining. (A) 1+, generally weak membrane or cytoplasmic staining in >10% of tumor cells; (B) 2+, moderately stained membrane or cytoplasm in >10% of tumor cells; (C) 3+, strongly stained membrane or cytoplasm in > 10% of tumor cells.

Fig A2.

Trop-2 IHC scoring vs. PFS (n = 48). (A) Kaplan-Meier plot for 4 separate groups with the 0 (n = 2) and 1+ (n = 4) combined, including 21 patients who did not have a specimen for Trop-2 analysis. (B) Kaplan-Meier plot comparing PFS for patients with specimens stained 0 and 1+ combined vs. those with 2+ and 3+ Trop-2 staining combined.