We present a case of BAP1-deficient breast cancer arising in a patient with BAP1 cancer syndrome. The association of BAP1 cancer syndrome with breast cancer has proven difficult to demonstrate, with cohort studies yielding inconclusive results [6, 7, 10]. These studies were limited by small numbers of participants, which is understandable considering the rarity of BAP1 cancer syndrome. Breast cancer, occasionally with early onset has been described in several families, most notably in the study by Popova and colleagues, which was the first to report the association of BAP1 germline pathogenic variants with RCC . The BAP1-positive proband in one of their families developed bilateral early onset breast cancer, in addition to several RCCs. Her half-sister, who was also a carrier of the familial BAP1 variant developed early-onset breast cancer as well. Interestingly, authors presented evidence for loss of heterozygosity at the BAP1 locus in the proband’s breast tumor. However, the overall incidence of breast cancer in their BAP1-positive families was deemed insufficient to include it in the BAP1 cancer syndrome spectrum. BAP1 germline variants are rare in patients with breast cancer undergoing genetic testing for hereditary cancer . When testing for somatic variants, BAP1 inactivation was seen in less than 1% of breast cancer cases, suggesting it is not a major breast cancer gene . A study involving 760 unselected Australian breast cancer patients detected loss of immunohistochemical BAP1 expression indicating biallelic inactivation in three patients (0.4%) . One of them had previously developed a breast cancer and another also had cutaneous melanoma, but patients were not screened for germline variants.
One of the intriguing aspects of BAP1-deficient tumors is their variable prognosis depending on tumor type. There is evidence that mesothelioma is less aggressive when it arises in the setting of BAP1 cancer syndrome, whereas the effect of somatic BAP1 variants on mesothelioma prognosis is minimal . BAP1-deficient uveal melanoma has a poor prognosis, particularly in cases due to somatic inactivation . BAP1 aberrations predispose to an aggressive type of RCC, with worse overall survival . The analysis of transcriptome profiling data sets indicates that lower BAP1 expression imparts an unfavorable prognosis for breast cancer patients . Some BAP1-associated tumors appear to have a poor prognosis due to their unresponsiveness to conventional therapy . Therapeutic approaches targeting the role of BAP1 in cancer are currently under investigation, with on-going clinical studies assessing the effectiveness of PARP, EZH2 and HDAC inhibitors . Immunotherapy with immune checkpoint inhibitors (ICI) elicits a response in 20–30% of unselected mesothelioma patients but only 5–15% of unselected patients with uveal melanoma . Studies in peritoneal mesothelioma and RCC patients have shown BAP1-related tumors have an inflammatory microenvironment, with increased immune cell tumor infiltration and PD-L1 expression [11, 17]. Figueiredo et al. recently demonstrated BAP1-deficient uveal melanomas exhibit an immunosuppressive environment, with upregulation of regulatory T-cells leading to inhibition of cytotoxic T-cells and proposed anti-CD38/anti-CD74 compounds could be used in combination with ICI to increase the response rates to therapy .
Most breast cancers appear to be intrinsically resistant to treatment with immunotherapy with only a minority of patients responding to monotherapy with ICI . Selecting patients most likely to respond is challenging. PD-L1 immunohistochemistry can produce inconsistent results, whereas TMB is relatively low in breast cancer. Additional biomarkers including variants in DNA damage repair genes are therefore being pursued . As our patient’s TNBC expressed PD-L1, a combination treatment with ICI and a taxane was initiated. Even after discontinuation of nab-paclitaxel, she remained in remission with maintenance atezolizumab. To our knowledge, this is the first report of a patient with BAP1-deficient breast cancer responding to immunotherapy. We hypothesize that the tumor’s immunophenotype and the patient’s complete and durable response to ICI might be the result of BAP1-deficiency, making BAP1 alterations a putative biomarker for response to immunotherapy in breast cancers that harbor them. A single case of a BAP1-defficient breast cancer responding to atezolizumab is by no means evidence of a role for BAP1 inactivation in response to immunotherapy, but we believe our findings are interesting and warrant further investigation.
Our report implies there is an association between BAP1 cancer syndrome and breast cancer. Additional epidemiological studies are needed to clarify the risk as this has important implications for patient management. Also, possible genotype–phenotype associations need to be explored. At the moment, breast cancer screening is offered to BAP1 carriers on the basis of personal risk factors and family history. Even if the risk of developing breast cancer is low or moderate for most patients, screening might be advisable if BAP1-deficient breast cancers are confirmed to be associated with poor prognosis and/or warrant specific therapeutic approaches.