Abstract
Background/Aim: The prolactin receptor (PRLR) is implicated in the tumorigenesis of breast and prostate cancers where it drives cell proliferation, survival, and migration. LFA102 is a humanized monoclonal antibody against PRLR with promising preclinical antitumor activity. To determine the maximum tolerated dose or a recommended dose, and to delineate the pharmacokinetic profile of LFA102 in Japanese patients, we conducted a phase I study. Patients and Methods: LFA102 was intravenously infused every 4 weeks to patients with advanced breast or castration-resistant prostate cancer, and the dose increased from 3 to 40 mg/kg. Results: Fourteen patients were treated, and toxicities were reported in 9 (64%) patients. They were all grade 1 or 2, and the most frequently observed toxicity was nausea (3 patients, 21%). No dose-limiting toxicities were observed. LFA102 did not show antitumor activity as a single agent. Conclusion: Treatment with LFA102 was well tolerated.
Prolactin mediates its physiological effects through interactions with the prolactin receptor (PRLR). Following binding to PRLR, prolactin activates a wide array of downstream signals, including JAK2/STAT5, JAK1/STAT3, SRC, and FAK pathways leading to PI3K/AKT and RAF/MEK/ERK activation. These signaling cascades trigger cell proliferation, survival, migration, differentiation, and angiogenesis (1). Preclinical data show that mammary gland- or prostate-specific expression of prolactin in transgenic mice increases the incidence of breast and prostate tumors, respectively (2-4). In the clinic, PRLR overexpression is observed in many malignancies, including breast, prostate, ovarian, colorectal, and pancreatic cancers (5-8). Indeed, 95% of breast and prostate cancers are PRLR-positive (9, 10) and high levels of prolactin in the blood have been correlated with increased risk and poor prognosis in breast cancer (11-13). In prostate cancer, prolactin expression has been correlated with phosphorylation of Stat5 (a key mediator of prolactin signaling), high Gleason scores and unfavorable prognosis (14, 15).
Prolactin protein is synthesized and secreted from the pituitary gland and from extrapituitary sites including breast and prostate tissues. The overexpression of PRLR in breast and prostate cancers stimulates their growth and contributes to their aggressiveness by autocrine and paracrine mechanisms (3, 16, 17). Furthermore, endogenous prolactin increased the constitutive tyrosine phosphorylation of HER2 in a breast cancer cell line, leading to enhanced growth through the RAS-MAPK pathway (18). Therefore, targeting the prolactin signaling pathway is an attractive endocrine therapeutic strategy.
LFA102 is a humanized IgG1/ĸ monoclonal antibody (mAb) against PRLR. The antibody neutralizes all prolactin-induced intracellular signaling pathways examined in vitro, including those mediated through Stat5, Akt, and Erk1/2. Furthermore, LFA102 induces antibody-dependent cell-mediated cytotoxicity in vivo, and inhibits the growth of prolactin-dependent cell lines engineered to express human PRLR and of mammary tumors induced by dimethylbenz[α]anthracene (19). LFA102 also blocks prolactin-induced proliferation in human breast cancer cell lines, including MCF7 and T47D (19).
In the present study, we aimed to determine the maximum tolerated dose (MTD) or a recommended dose of LFA102, and to investigate safety, pharmacokinetics, and preliminary evidence of anti-tumor activity in Japanese patients. We therefore conducted a phase I open-label multi-center dose-escalation study in patients with advanced breast cancer or castration-resistant prostate cancer (CRPC). This study was conducted in parallel with a global phase I study in the US and European countries in patients with PRLR-positive metastatic breast cancer or CRPC (20).
Patients and Methods
Eligibility criteria. Patients with histologically/cytologically confirmed advanced breast cancer or CRPC, age ≥18 years, Eastern Cooperative Oncology Group (ECOG) performance status 2 or better, adequate bone marrow, liver and renal functions, and no anticancer drug therapies for ≥4 weeks (6 weeks for nitrosourea, or mitomycin-C) were enrolled.
Prostate cancer patients must have had detectable metastasis and should have been castrated either surgically via orchiectomy or chemically via the use of luteinizing hormone-releasing hormone (LH-RH) agonists or antagonists. Progressive disease must also have been documented. For breast cancer, progression following the last line of standard treatment and at least one measurable lesion as defined by Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 were required.
The main exclusion criteria were untreated and/or symptomatic metastatic central nervous system disease, other active malignancies or clinically significant diseases, prior treatment with any anti-PRLR antagonist, major surgery within 28 days, radiotherapy within 2 weeks, chronic steroid therapy of more than 10 mg/day prednisone, and pregnant or nursing women.
Study treatment and design. LFA102 was administered once every 4 weeks (28-day cycle) as an intravenous infusion over 2 h. Treatment was continued until disease progression, unacceptable toxicity, or patient withdrawal.
A Bayesian logistic regression model (BLRM) with an overdose control principle was used to support the decision of dose escalation and determination of the MTD or a recommended dose (21, 22), and each cohort of patients consisted of 3 to 6 evaluable patients. Provisional dose levels were 3, 10, 20 and 40 mg/kg, and the starting dose of 3 mg/kg was based on information on the highest non-severely toxic dose in cynomolgus monkeys. Dose-limiting toxicities (DLTs) observed during the first cycle of treatment with LFA102 were considered at dose escalation.
The dose was escalated until the MTD was reached. The MTD was defined as the highest dose not causing medically unacceptable DLTs in more than 33% of the treated patients in the first cycle of the treatment. The recommended dose was determined in consideration of all clinical data obtained in the entire study. The study was approved by the ethics committees of each participating institution and conducted in accordance with the International Conference on Harmonization Good Clinical Practice Guidelines as well as the Declaration of Helsinki. All patients provided written informed consent.
Definition of DLTs. Toxicities were graded according to the National Cancer Institute's Common Terminology Criteria for Adverse Events (CTCAE) version 4.03. DLT was defined as grade 4 neutropenia for >5 consecutive days, febrile neutropenia, ≥ grade 3 thrombocytopenia with clinical bleeding or requiring platelet transfusion, ≥ grade 4 thrombocytopenia, ≥ grade 3 increase in serum creatinine or total bilirubin, grade 2 total bilirubin elevation in combination with grade 3 AST or ALT elevation, grade 3 elevation of AST/ALT for >7 days, grade 4 AST/ALT elevation, symptomatic serum amylase/lipase elevation or pancreatitis, ≥72 h of grade 3 fatigue, ≥ grade 3 other non-hematological toxicities, and discontinuation from study drug administration (including dose delay of >4 weeks from the next scheduled treatment).
Patient characteristics.
Concomitant treatment. Patients with prostate cancer under treatment with LH-RH analogues or antagonists were required to continue these treatments to maintain the castrated status. Daily use of 10 mg prednisone (or equivalent) or lower dose was permitted. Patients were not allowed to take medications that could affect prolactin level, including drugs such as butyrophenones, bromocriptine, clomipramine, cabergoline, terguride, phenothiazines, risperidone, and thioxanthenes (23, 24).
Assessments. Safety was assessed by grading toxicities based on the CTCAE. For efficacy assessment, contrast-enhanced CT or MRI imaging was performed every 8 weeks and efficacy was evaluated according to the RECIST version 1.1 in patients with breast cancer. In patients with CRPC, contrast-enhanced CT or MRI imaging was performed every 12 weeks, and PSA was measured every 4 weeks. Efficacy was evaluated using the Prostate Cancer Clinical Trials Working Group (PCWG2) response criteria (25).
Toxicities.
Pharmacokinetics. For pharmacokinetic investigation, blood was obtained at baseline, 1, 2 (end of infusion), 2.25, 4, 8, 24, 48, 72, 168, 240, 336, and 504 h after the start of infusion of cycle 1, as well as before infusion of each cycle. Serum concentrations of free LFA102 were measured using a validated assay method with a LLOQ of 200 ng/ml.
Results
Demographic and background characteristics. A total of 14 patients were enrolled and treated with LFA102 at doses ranging from 3 to 40 mg/kg (Table I). Among seven patients with breast cancers, one patient had triple-negative breast cancer (negative for estrogen receptor, progesterone receptor, and HER2) and one patient had breast cancer positive for the three markers. For metastatic disease, all patients had been treated with hormonal therapy except for the patient with triple-negative breast cancer. Patients with breast cancer had been heavily treated with chemotherapy (Table I).
The median duration of treatment was 12.35 weeks (range=4.0-45.1 weeks) across all doses, and six patients (43%) were treated for ≥14 weeks. No anti-tumor activity was evident.
Safety. LFA102 was safe and well tolerated. Of 14 patients, drug-related toxicities were reported in 9 (64%) patients, and the most frequently observed toxicities were nausea (3 patients, 21%) followed by dry skin and peripheral edema (2 patients each, 14%; Table II). All toxicities were grade 1 or 2 in severity. The incidence of toxicities was generally comparable across all LFA102 doses.
Most of the adverse events regardless of the relationship with LFA102 were grade 1 or 2. Grade 3 or 4 adverse events were reported in 3 patients (21%), and composed of increased amylase level, dyspnea, lymphocytopenia, peripheral motor neuropathy and tumor pain occurring in one patient each. None of these grade 3 or 4 adverse events were suspected to be related to LFA102. Two patients (14%) reported serious adverse events. These were nausea, vomiting, urinary tract infection, and pathological fracture. All of these events were grade 2 in severity and none of them were considered to be related to LFA102.
Overall, one patient (7%) discontinued the study drug due to nausea and vomiting; neither illness was considered to be related to the study drug. No dose adjustment or interruptions were precipitated by adverse events. No DLT was observed up to 40 mg/kg, and the MTD was not reached.
Pharmacokinetics. Exposure to LFA102 increased almost dose-proportionally with dose escalation, and inter-individual variability of exposure to LFA102 at each dose was small with the coefficients of variation of AUC and Cmax ranging from 9 to 24% (Table III). Median half-life at each dose ranged from 6.6 to 12.1 days, which is relatively short for an antibody drug.
Pharmacokinetic parameters.
Discussion
Toxicities associated with LFA102 were mild and no grade 3 or greater toxicities were observed; the most common toxicity was nausea. The dose of LFA102 was increased to 40 mg/kg in this study without observations of DLTs, and the MTD was not reached. Similarly, in a parallel phase I study of LFA102 conducted in the US and Europe in which the dose was escalated to 60 mg/kg, no DLTs were observed and the MTD was not reached (20).
A dose-proportional increase in exposure to LFA102 was observed with dose escalation, and the inter-individual variability of pharmacokinetics was small. Exposure to LFA102 was also dose-proportional in the parallel phase I study (20), and there was no evidence of inter-ethnic differences in pharmacokinetics between Japanese patients and those in the US and Europe.
Dose escalation was halted because no responses were observed either in this study or in the parallel phase I study. Preclinical evidence suggests that prolactin signaling has an important role in breast and prostate cancer, and LFA102 treatment did elicit tumor shrinkage in an in vivo breast cancer model (19). The situation in the clinic is clearly different, since LFA102 seemed to have no efficacy as a single agent in patients with advanced breast cancer or CRPC.
LFA102 was administered every 4 weeks in both phase I studies, and had a half-life of 7 to 12 days, which is relatively short for an antibody drug. We infer that more frequent dosing of LFA102 might be necessary for antitumor efficacy. However, in a preclinical in vivo breast cancer model, LFA102 suppressed tumor growth with regression observed in 20% of tumors, and phosphorylation of Stat5 in cancer cells was inhibited at an LFA102 serum concentration of 70 μg/ml or higher (19). In our phase I study, the Cmax of LFA102 was much higher than this which, given its half-life, should have been sufficient to suppress Stat5 phosphorylation in most patients during the treatment courses.
In the parallel phase I study, serum prolactin level was increased in a dose-dependent manner by LFA102 treatment, suggesting prolactin signaling was suppressed in patients (20). However, the pituitary gland is the main source of prolactin, and it is unclear whether LFA102 suppresses prolactin signaling in the tumor tissue. It is possible that the distribution of LFA102 to cancer tissues in patients was limited.
Potential explanations for the lack of efficacy might include the possibility that prolactin is not an oncogenic driver in human breast or prostate cancer, that other pathways are activated in order to bypass the effect of PRLR inhibitors, or that prolactin-independent signaling pathways are upregulated in order to maintain tumors. Although many preclinical studies have demonstrated that PRLR activation can promote cancer cell proliferation and survival (1, 16, 17), the clinical effect of cancer cell-derived prolactin on patient outcome is controversial in recent reports. High PRLR expression was associated with a shorter time to bone metastasis in one study of patients with breast cancer (26), whereas other reports showed that it was associated with better relapse-free survival and longer distant metastasis-free survival (27, 28). Recently it has also been reported that PRLR is a marker of good prognosis in triple-negative breast cancer patients who underwent surgery (29). In line with these reports, expression/activation of Stat5a, an effector molecule in prolactin signaling, was positively correlated with an increased level of histologic differentiation in breast cancer tissues and with a favorable prognosis (30). Thus, whereas prolactin may contribute to the initial development of breast cancer, it may also have context-specific roles in restricting the metastatic potential of certain tumors (31).
Furthermore, prolactin potently inhibits growth factor-induced proliferation of breast cancer cells (32). Preclinical experiments indicate that blocking prolactin signaling leads to activation of mitogen-activated protein kinase and transforming growth factor-β/Smad signaling pathways (33), and concomitant expression of prolactin and TGF-β receptors is associated with a less aggressive phenotype and favorable patient outcome (34). Potential agonistic as well as antagonistic effects of prolactin signaling demonstrate the need for further biological study.
Signals from PRLR crosstalk with other signaling pathways including those activated by estrogens, androgens, and growth factors including the IGFs and TGF-α (16, 27. This suggests that LFA102 may be particularly effective in combination with other hormonal therapies or molecular targeted drugs. For example, LFA102 potentiated the efficacy of an aromatase inhibitor, letrozole, in a preclinical breast cancer model (19). Autocrine prolactin can activate the oncogenic receptor HER2, and in HER2-positive breast cancers this is associated with increased proliferation and metastasis. Interestingly, the combination of trastuzumab and inhibition of PRLR signaling blocks cellular proliferation (18, 35). In another example, a PRLR antagonist augmented the cytotoxic effects of doxorubicin and paclitaxel in preclinical experiments, whereas it was ineffective as a single agent (36). LFA102 in combination with other agents that inhibit the above pathways might therefore be a potential strategy to pursue.
In conclusion, LFA102 up to 40 mg/kg was well tolerated and showed an acceptable safety profile in Japanese patients. No anti-tumor activity was observed in patients with advanced breast cancer or CRPC when LFA102 was used as a single agent. More studies are needed to understand the role of PRLR-driven signaling cascades in tumor growth, especially with regard to cross-talk between the PRLR pathway and oncogenes, growth factors and hormones. It will also be necessary to develop accurate biomarkers to identify patient subgroups that could benefit from PRLR blockade.
Footnotes
Authors' Contributions
H.M. enrolled and treated patients, interpreted data, and wrote the manuscript; Y.A. and K.T. enrolled and treated patients, interpreted data, and reviewed the manuscript; T.T. reviewed the pharmacokinetic part of the manuscript; R.I. supervised the study concept and design, interpreted data, and reviewed the manuscript.
This article is freely accessible online.
Conflicts of Interest
H.M. received research grants and honoraria from Bayer Yakuhin, Boehringer Ingelheim, Bristol-Myers Squib, Chugai Pharmaceutical, DaiichiSankyo, Eisai, Kyowa-Kirin, Merck Serono, MSD, Novartis, Ono Pharmaceutical, Pfizer, Sanofi, Takeda Pharmaceutical, and Taiho Pharmaceutical, Eli Lilly; research grants from Asahi-Kasei Pharma, Astellas Pharma, Nippon Shinyaku, Yakult Honsha, CSL, Behring, and Nippon Kayaku; honoraria from Celgene, Ohtsuka Pharmaceutical, Shire Japan, Genomic Health, and Abbvie. Y.A. received research grants and honoraria from Chugai Pharmaceutical Co., Ltd., Kyowa Hakko Kirin Co., Ltd., Nippon Kayaku Co., Ltd., Yakult Honsha Co., Ltd., Mochida Pharmaceutical Co., Ltd., Ono Pharmaceutical Co., Ltd., Taiho Pharmaceutical Co., Ltd., and Daiichi Sankyo Company, Ltd.; research grants from Eisai Co., Ltd.; and honoraria from Eli Lilly Japan K.K., Bristol-Myers Squibb, Sawai Pharmaceutical Co., Ltd, Tsumura & Co., Shionogi & Co., Ltd, Novartis Pharma K.K., and Bayer Holding Ltd. outside the submitted work. K.T. received research funding from Pfizer, Daiichi-Sankyo, Eli Lilly, Chugai, AstraZeneca, MSD and Novartis. T.T. is an employee of Novartis Pharma K.K. R.I. is an employee of Novartis Pharmaceuticals Corporation.
- Received July 21, 2020.
- Revision received July 27, 2020.
- Accepted July 28, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.