Clinical and Practical Implications of the Breast Cancer Index


A review of when to use the Breast Cancer Index for patients with HR+ breast cancer when deciding on endocrine therapy duration.

Mark Pegram, MD: Tell us the highlights about how that Breast Cancer Index [BCI] has been tested, from the Stockholm study to MA.17 to Trans-aTTom to IDEAL and beyond.

Lee Schwartzberg, MD, FACP: The other part of the BCI is the molecular grade index, which is a series of genes that also measures proliferation, among other things. It’s a different set of genes than we see in any of the other classifiers. There’s some overlap across all the classifiers, but this is a different set as well.

Mark Pegram, MD: There are 5 genes different from HOXB13 that don’t have ER [estrogen receptor], PR [progesterone receptor], or Ki67.

Lee Schwartzberg, MD, FACP: Interestingly, they’re combined to give a predictive value, and it’s a continuous variable that can be separated into a low-risk group and a high-risk group. What’s important about BCI is that it’s looking at and has been validated in multiple tumor sets and multiple clinical trials from a retrospective, prospective approach—much like Oncotype DX took—to look at the risk of recurrence from years 5 through 10. That’s the unique aspect of BCI that the other genomic classifiers aren’t as good at doing. Looking at late relapse and specifically using a landmark analysis starting at year 5 through year 10, we get a prognostic value, which clearly separates patients into low-risk or high-risk groups. You basically get that determination from looking at the 5 genes plus the H/I [HOXB13/IL17BR] index. That’s the first part of the test.

The second part of the test, which has perhaps more clinical value, is predictive value. In this case, predictive value for the use of extended adjuvant endocrine therapy as opposed to the use of chemotherapy, which we see with the other classifiers and certainly with Oncotype DX. There have been multiple data sets that have been examined, and the results have been pretty consistent, starting with the Stockholm study, which continues to generate lots of information. It’s over 25 years old. These patients got tamoxifen or didn’t, and all the blocks were kept from these patients for a very long time, so we’re looking truly at an experimental group with almost 100% follow-up. That was the first validation of the BCI.

Then [there were] randomized trials. We mentioned MA.17, which was probably the first large-scale extended endocrine therapy trial, which showed the prognostic benefit of BCI and, importantly, that patients who had a high H/I score had benefit from that additional 5 years of letrozole compared with those who had a low score who didn’t have a significant benefit there. That was the key. It has been studied in Trans-aTTom and in IDEAL, and showed 7 years vs 10 years in IDEAL. You could identify the patients who would benefit for 10 years vs 7 years with the H/I score as well. We have a large body of data with BCI that give us great prognostic and predictive information to use when we’re making that decision with our patients at around the 5-year mark regarding whether they should continue.

Mark Pegram, MD: That collection of translational trials is a nice collection because it has examples of 5 years with tamoxifen followed by 5 years of an AI [aromatase inhibitor], like MA.17. It has an example of 5 years of tamoxifen followed by 5 more years of tamoxifen in Trans-aTTom. And IDEAL had a subgroup of patients that had 5 years of an AI followed by extended adjuvant AI for 2 to 5 additional years as well. All of those trials indicated an advantage in terms of the therapeutic advantage for the BCI-high group but not the BCI-low group. That’s a strength of these big data sets in translational fashion. Finally, I’ll note that the NCCN [National Comprehensive Cancer Network] and ASCO [American Society of Clinical Oncology] guidelines committees have both recognized this 2-gene ratio now based on the strength of all of these translational studies, and they’re now listed in the guidelines for consideration of application in the clinic.

Lee Schwartzberg, MD, FACP: I’ve used the BCI for several years now, and it’s typically done the same way the other genomic classifiers are: on the primary tumor. I know and our audience should know—and probably knows—the fact that by law, tumor blocks are required to be saved for 10 years and accessible for 10 years. It should be relatively easy to find these blocks, even if you’re thinking about it at year 4 and want to do the test and then have an informed decision with your patient by giving them this extra tool, which is very valuable on top of the clinical features and patient preference to make that decision. This is 1 more valuable piece of information that helps patients make the best decision with their doctor.

Mark Pegram, MD: One final point that we should make is the issue of clinical parameters that have prognostic significance vs biomarkers that have predictive value for benefit from extended endocrine therapy. This came to light at ASCO [American Society of Clinical Oncology Annual Meeting] 2022. There was a nice paper presented using a clinical classifier called CTS5 [clinical treatment score post-5 years], which is a combination of lymph node status, age, tumor size, and tumor grade. They applied that metric to the IDEAL trial, and they had data from the Breast Cancer Index 2-gene ratio. Those studies had already been conducted for the IDEAL trial, and they showed that they predicted for different things. The clinical CTS5 score was only prognostic; it didn’t predict benefit from extended adjuvant endocrine therapy in IDEAL, whereas the 2-gene ratio BCI biomarker was able to retain its predictability in IDEAL in all of the arms of that study. It underscores and highlights that it’s a different measure that’s a factor that predicts response to therapy, not purely a prognostic risk score like CTS5 or other clinical parameters that are commonly used in clinical practice. It really is unique.

Lee Schwartzberg, MD, FACP: Yes, and that’s definitely the clinical utility strength of the assay. Because while knowing prognosis and advising patients on it is useful, unless there’s something you can do about it or not do if they don’t need it, you’re left in a difficult position. Having this predictive biomarker is extremely valuable in making this decision for extended adjuvant endocrine therapy.

Mark Pegram, MD: I should mention the biomarker interaction key value was statistically significant in the case of the BCI but was insignificant for the CTS5 prognostication clinical parameter score in that paper at ASCO [Annual Meeting] 2022.

Transcript edited for clarity.

Related Videos
Reshma Jagsi, MD, DPhil, Emory University
Jane L. Meisel, MD
Kari Wisinski, MD, Endowed Professor of Hematology and Oncology, chief, Division of Hematology, Medical Oncology, and Palliative Care, Department of Medicine, University of Wisconsin (UW) School of Medicine and Public Health; associate director, Clinical Research, co-lead, Breast Cancer Disease Oriented Team, UW Carbone Cancer Center
Corey Cutler, MD, MPH, and Hana Safah, MD, experts on GvHD
Corey Cutler, MD, MPH, and Hana Safah, MD, experts on GvHD
Mitchell S. Cairo, MD, and Sergio Giralt, MD, experts on veno-occlusive disease