Currently, the American Cancer Society recommends that women who are at higher risk for developing breast cancer have annual dynamic contrast-enhanced MRI (DCE-MRI) and mammography screening, starting at 30 years. At the 2017 San Antonio Breast Cancer Symposium, Olufunmilayo Olopade, MD, the Walter L. Palmer Distinguished Service Professor of Medicine and Human Genetics at University of Chicago Center for Clinical Cancer Genetics, presented data from a study conducted jointly there and in collaboration with the University of Washington, which evaluated the use of semiannual (i.e., every 6 months) DCE-MRI screening in women who had been genetically confirmed to have a higher chance for developing breast cancer (Poster Session P4-02-10).
Regarding these findings, Olopade stated: “Because of intensive surveillance and adoption of evidence-based interventions to reduce risk, the majority of high-risk women in this study—most of whom had highly penetrant genetic mutations—have not developed breast cancer.”
Poster Session P4-02-10
From 2004 to 2016, a prospective cohort was assembled at the University of Chicago that consisted of women who were deemed to be at high risk for developing breast cancer and were undergoing semiannual DCE-MRI and annual mammography screenings.
Eligibility criteria included a greater than 20 percent lifetime risk of developing breast cancer and/or been assessed for pathogenic mutations with a cancer gene panel that includes ATM, BLM, BRCA1, BRCA2, BRIP1, CDH1, CHEK2, PALB2, TP53, and other DNA repair pathway cancer susceptibility genes. When somatic mutations were observed in screened tumors, subsequent evaluations were performed on DNA obtained from formalin-fixed paraffin embedded samples using the UW-OncoPlex cancer gene panel.
After physician evaluation and genetic counseling, participants were scheduled to undergo a clinical breast examination with DCE-MRI scans every 6 months and digital mammograms every 12 months. The patients in this study also had genetic tests that evaluated the samples for 12 genes associated with having a predisposition to breast cancer. Patients identified as having high genetic risk mutations who completed 5 years of the study protocol were offered continued screening. The enrollment for this study was closed in December 2016 and a new protocol is currently in development to test whether time on the MRI machine could be dramatically shortened to lower costs.
When asked about the contrast agents used in this trial, Olopade replied, “Investigators in our program are continually investigating different gadolinium-containing complexes to continue to advance the science and make MRI imaging cheaper and safer for patients who stand to benefit from novel imaging technologies.”
Investigators recruited 305 patients for inclusion in this study, however, 10 patients dropped out of this trial before their first screening. For the remaining 295 participants, the mean age at entry into this study was 43.3 years. The largest number of patients included Caucasian (85.4%) followed by African-American (11.5%). The menopausal status for the participants was as follows: pre-menopausal—47.5 percent; post-menopausal with bilateral salpingo-oophorectomy (BSO)—31.9 percent; post-menopausal no BSO—13.9 percent; status missing—6.8 percent.
Of the patients in this study, 43.4 percent had wild-type (i.e., non-mutated) status for all genes tested, 53.2 percent were mutation carriers, and 3.4 percent did not have testing done. The most commonly encountered germline deleterious mutations were: BRCA1—25.4 percent; BRCA2—20.7 percent; CHEK2—3.4 percent; CDH1—1.4 percent; PALB2—1.0 percent. Interestingly, two patients had multiple mutations present. The first of these participants had mutations to both the BRCA1 and ATM genes, while the other had mutations for their BRCA2 and BRIP1 genes.
A clear majority of the participants in this study had no history of breast or ovarian cancer (78.3%). Of those that did, 18.3 percent had breast cancer only, 1.4 percent had ovarian cancer only, and 2.0 percent had both breast and ovarian cancer.
Over the course of the study, the participants received 2,111 DCE-MRIs, for an average of about seven per patient, and 1,223 mammograms, or roughly four per participant. All cancers detected during the study were smaller than 1 cm.
“Semiannual DCE-MRI performed well for early detection of invasive breast cancer in high-risk women, accomplishing the ultimate goal of breast cancer screening—detecting node-negative, invasive tumors less than 1 cm,” Olodape noted.
The sensitivity for this imaging method was 88.2 percent, while the specificity was 96.8 percent. It should be noted that DCE-MRI imaging obtained at the 6-month interval allowed the detection of six invasive cancers. DCE-MRI did miss one high-grade ductal carcinoma in situ (DCIS) (0.5 cm) and an intermediate-grade DCIS found in a sample from a patient who had prophylactic mastectomy.
During this study, 17 patients were diagnosed with breast cancer; of these, four had DCIS and 13 were early-stage breast cancer. These 17 cancer patients were continuously followed for a median period of 5.3 years with a range of 0.5-11 years. Fifteen of these participants had mutations present: BRCA1—11; BRCA2—3; CDH1—1. The mean tumor size found in this study was 0.61 cm with sizes ranging from 0.1 to 1.0 cm. None of the patients had interval invasive tumors or lymph node involvement.
Somatic mutation analyses were somewhat limited, as a majority of patients had insufficient amounts of DNA available for testing. Genetic analysis of the tumors revealed that all had mutations to TP53.
For patients who withdrew from the study within 5 years (n=121), the following reasons were cited most often: health insurance or other payment issues—28.1 percent; reasons unknown or lost to follow-up—19.0 percent; prophylactic mastectomy—16.5 percent; inconvenient/moved away/changed care provider—9.1 percent; pregnancy—9.1 percent.
“These scans performed especially well in BRCA1 mutation carriers, who are at risk for aggressive subtypes of breast cancer,” Olopade observed. She further noted that “mammograms remain important for most women; however, for high-risk patients who are getting a DCE-MRI every 6 months, annual mammograms could probably be eliminated. For this group of younger women at significantly elevated risk, especially those with a BRCA1 mutation, we strongly support intensive surveillance, including studying whether a DCE-MRI every 6 months would be beneficial.”
In summing up the most significant findings in their study, Olopade explained, “This is a single institution study and we learned a lot about women who did not want to have their healthy breasts removed. They were highly motivated and the study demonstrated, for the first time, that aggressive breast cancers can be caught early, without excessive recalls or biopsies.
“We hope that by showing the effectiveness of using DCE-MRI for early breast cancer diagnosis, there will be increased interest in helping women get insurance coverage for prevention services. DCE-MRI and prevention services could be optimized if we can identify women who need the services the most, as cost/insurance coverage was the most-cited reason for withdrawal from our study,” Olopade stated.
“For many patients who have been confirmed to be mutation carriers of BRCA1, this diagnosis can be quite stressful; many patients opt to have prophylactic mastectomies in order to prevent the development of breast cancer,” Olopade noted. “When we undertook this study, we sought to give these patients options other than life-changing prophylactic mastectomies for dealing with their potential future breast cancer development.
“We are also glad to report that all of the breast cancer patients identified in our study are currently alive and free of systemic disease because we continue to make progress in treating breast cancer. It is important for us to reassure patients at high genetic risk that significant progress has been made through the brave women with BRCA1 and BRCA2 mutations who came forward to advance this research, and now we need to test whether our approach can be replicated in a multicenter national study,” she concluded.
Richard Simoneaux is a contributing writer.