Designing an immunofluorescence-based biomarker assay for detecting Rb and phospho-Rb expression in HR+/HER2- breast cancer
Introduction
Cyclin-dependent kinases 4/6 inhibitors (CDK4/6i) combined with endocrine therapy are the standard of care in hormone receptor-positive/human epidermal growth factor receptor-2-negative breast cancer (HR+/HER2- BC). However, almost 50% patients are resistant to this therapy. Therefore, there is an unmet clinical need for biomarkers to predict treatment response to spare non-responders from toxicities and financial burden. The aberrant or loss of Rb is common in patients with CDK4/6i resistance as evidenced from ctDNA and immuno-histochemistry analyses on tumor biopsies. Further, the reduction in phospho-Rb (pRb) levels during therapy is also associated with CDK4/6i response. However, repeat tumor biopsies are often not feasible and are invasive. Therefore, we aimed to develop a multiplex immunofluorescence (IF) assay to predict tumor response to CDK4/6i [palbociclib, ribociclib, and abemaciclib] using circulating tumor cells. Here, we report the selection of various internal controls for various protein markers in the assay and optimization of IF protocol.
Methods
The CTC detection biomarkers were DAPI, cytokeratin (CK), and CD45. The treatment response biomarkers were Rb and pRb for efficacy of CDK4/6i and estrogen receptor α (ER) for endocrine therapy efficacy. Parental cells (P) of two HR+/HER2- cell lines, MCF7 and T47D, were used as reference for CK, Rb, pRb, and ER expression. The endocrine resistant (estrogen deprivation-resistant/EDR and fulvestrant-resistant/FulR), CDK4/6i resistant (palbociclib-resistant/PalboR), and endocrine+CDK4/6i-resistant (EDR/palboR) derivatives of MCF7 and T47D cells were assessed for expression of Rb, pRb, and ER. In addition, Rb1 KO cell models of MCF7 and T47D as well as BT549 and MDA-MB-468 were assessed for serving as negative controls for Rb and pRb. Leukocytes from healthy subjects were evaluated for expression of CK and CD45. Antibodies for individual protein markers were tested on various candidate positive and negative control cell lines to test the selectivity and specificity of the antibodies. The slides were imaged using Leica SP8 STED confocal microscope. The concordance of Rb and pRb IF staining intensity was evaluated with the western blotting (WB) results on the same samples and the level of agreement was assessed using a Bland-Altman plot of difference.
Results
Our initial approach utilized indirect IF staining against Rb and pRb. The analytical range for mean Rb and pRb staining intensity was 0.744 – 46.779 and 1.001 – 52.919, respectively. The same primary antibodies were also used for WB. The level of bias for Rb and pRb was 0.086 and 0.157 on the Bland-Altman plot. The upper and lower limits of agreement were 0.398 and -0.226 for Rb and 0.728 and -0.415 for pRb. Most data points were relatively close to the bias line, suggesting that the difference between the two methods was acceptable. Based on WB and IF data, we nominate T47D PalboR and MDA-MB-468 cells for serving as negative controls for Rb and pRb expression. MCF7 FulR and T47D EDR were found to be adequate negative control for ER expression. Leukocytes from healthy controls were positive for CD45 and negative for CK, as expected. Out of 138 positive samples and 27 negative samples of various cell line models and biological replicates, the selectivity and specificity were 100% and 70%, respectively, for the Rb expression and 99% and 61% for pRb, respectively. To circumvent higher non-specific staining from indirect IF, we are evaluating and optimizing the use of directly conjugated primary antibodies
Conclusion
We have selected internal cell line controls for all biomarkers in our IF assay and are in process of selecting antibodies and conditions to optimize signal-to-noise ratio and specificity. Ongoing studies include automated quantitation using digital whole slide fluorescent imaging and multiplexing the IF assay.