Breast cancer is still a big threat to women’s health accounting for 30% of newly diagnosed cancers in great medical demand [
37]. It’s confirmed that prolactin receptor plays an important role in the development of breast cancer [
38]. Immunotherapy for the treatment of cancer get more attention recently. The BsAbs which could redirect the T cells to tumor cells and kill them showed to be a highly promising strategy both in pre-clinical and clinical sets [
11–14,19–21]. And here in our study, we generated a bispecific antibody targeting the PRLR and CD3 exhibiting a favorable activity both in vitro and in vivo. Agarwal
et al has reported a negative result of PRLR mAb LFA102 (a kind of humanized mAb binding with PRLR to inhibit the PRLR-mediated signaling transduction) for the treatment of metastatic PRLR positive breast cancer and metastatic castration resistant prostate cancer in phase I clinical trials (10). In our study, PRLR-DbsAb could enhance the cytotoxicity by more than 2 times relative to the top cytotoxicity observed for PRLR mAb (Fig.
2h). The dosage of PRLR-DbsAb needed for achieving 30% killing of the PRLR highly expressing cell lines is only 10 ng/ml (pM) while it is 1000 ng/ml for PRLR mAb, which is 100 times higher (Fig.
2f). Tumor-bearing mice were used to investigate the antitumor activity of PRLR-DbsAb treatment with intraperitoneal (i.p) injection because there is no differences in the peak plasma concentrations between i.p and i.v administration of antibody in mice [
39] and i.p administration is relatively easy to perform the in vivo experiment. In vivo results illustrated that 0.33 mg/kg PRLR-DbsAb had the comparable anti-tumor activity with PRLR mAb at a dosage of 3 mg/kg (Fig.
7c). The Fig.
2h experiment had shown from in vitro perspective that the anti-tumor effects of PRLR-DbsAb were not superimposed by the anti-tumor effects of CD3 and PRLR monoclonal antibody. The in vivo experiment would be more helpful to illustrate the effectiveness of PRLR-DbsAb using the mixture of both monoclonal antibodies as the control group, which will be reported in our future in-depth study. We used the NOD/SCID mouse model to assess the PRLR-DbsAb activity in vivo due to its unique advantage in evaluating candidate drug in human tumor transplanted animal systems. However, there is a main disadvantage of this model, that it is not able to provide human immune cells which usually have very limited half-life in the murine background [
40]. In addition, we injected huPBMCs with peritoneal administration, which could further exacerbate this problem. Taking these factors into consideration, we believe the full potential of PRLR-DbsAb is underestimated, especially at the thought of a great number of T cells circulating in the human body in clinical sets. PRLR is overexpressed in certain types of breast cancer as well as prostate cancer. Unlike another target for breast cancer HER2 (expressed in most normal epithelial cells), PRLR is only slightly expressed in normal breast tissues. In the treatment of advanced breast cancer, surgery is routinely performed to remove the potentially infiltrating breast tissue [
41,
42]. In Phase I clinical climbing trials (3-60 mg/kg), 73 patients were treated, including 53 at the highest dose level. Overall, LFA102 was safe and well-tolerated; the maximum tolerated dose was not reached, as there were no dose-limiting toxicities [
6]. These researches indicated minor severe side effects of PRLR immunotherapy. PD-L1 which is expressed on many cancer and immune cells, plays an important role in blocking the ‘cancer immunity cycle’ by binding to PD-1 and B7.1 (CD80), both negatively regulate T-lymphocyte activation. Binding of PD-L1 to its receptors suppresses T-cell migration, proliferation and secretion of cytotoxic mediators, and restricts tumor cell killing [
43,
44]. Triple negative breast cancer (TNBC) expressing PD-L1 participates in tumor immune escape by binding PD-1 to inhibit T cell activation [
45,
46]. Here, our in vitro experiments showed that inhibition of PD-1 enhanced PRLR-DbsAb killing of PRLR low expression TNBC cells. PD-L1 weakly expressed on cell line gradually unregulated with the prolongation of cell killing time mediated by T cell, suggesting that PD-L1 expression level in response to immunotherapy. PD-L1 expression is predictive of a clinical response to anti-PD-1/PD-L1 therapies [
47]. And there are extensive CD8+/PD-L1+ immune infiltration in breast cancer [
48]. It is noteworthy that a response to PD-1/PD-L1 therapy is strongly associated with immune cell with PD-L1 expression rather than tumor cell with PDL1 expression [
49]. Even more surprisingly in the study, PRLR-DbsAb significantly up-regulated PDL1 expression on CD4+ and CD8+ T cells compared to PRLR mAb treatment. The results of ex vivo experiments showed that the immunotherapeutic bispecific antibody PRLR-DbsAb may suppress the immune response by upregulating PD-L1 in lymphocyte besides PD-L1 in tumor. More in vivo and in vitro studies need to be conducted to investigate the role of PD-1/PD-L1 signaling pathways in the treatment of tumors with bispecific antibodies.