Preventing Tumor Progression in Women with High-Risk DCIS

Institution: University of California, San Francisco
Investigator(s): Thea Tlsty, Ph.D. -
Award Cycle: 2016 (Cycle 22) Grant #: 22OB-0032 Award: $956,774
Award Type: Translational Research Award
Research Priorities
Detection, Prognosis and Treatment>Innovative Treatment Modalities: search for a cure

Initial Award Abstract (2016)

Non-technical overview of the research topic and relevance to breast cancer: Ductal carcinoma in situ (DCIS), a ‘pre-cancer’, is also called ‘Stage 0’ or premalignant, which means that cancer cells are still inside the milk duct of the breast. Unfortunately, most women with DCIS feel that they have cancer because the word ‘carcinoma’ suggests cancer, and they often receive the same treatments as women with invasive breast cancer (IBC). About 61,000 women will be diagnosed with (pure) DCIS in 2016, in addition to the 246,660 women projected to be diagnosed with IBC (including DCIS combined with IBC). Most women with DCIS do well, only ~15% of them develop IBC over 10 years; but early identification is critical since IBC often spreads quickly. However, women who are diagnosed with DCIS by palpation (by feel) have a 2.1-2.4-fold higher risk of developing IBC compared to those diagnosed by mammography. This risk is even higher in underserved ethnic minorities. These women often have larger DCIS lesions with specific features including a stiffer tissue surrounding them (called the ‘stroma’ which acts like soil to cancer cells (‘the seeds’)). While it is important to ensure that these women at high risk receive full treatment, traditional assays cannot currently tell who they are. Therefore most women diagnosed with DCIS get the same type of surgery (mastectomy or partial mastectomy, i.e. full or partial breast removal) or partial mastectomy followed by radiotherapy and/or hormone drug therapy as women with IBC. This means that approximately 52,000 women will be treated unnecessarily with radiation, hormonal therapy, and/or mastectomy in order to prevent an IBC that may not, in fact, happen, causing unnecessary suffering and side effects in otherwise-healthy women and causing a healthcare burden.

The question(s) or central hypotheses of the research: We have identified biological changes (called ‘markers’) that help us start to predict future IBC in women diagnosed with pure DCIS. In this project, we hope to find more markers in breast tissue of women with palpable DCIS that will help us identify risk levels in women who are diagnosed with DCIS by palpation, and maybe for women with no palpable lesions. We will achieve this by comparing the levels of selected markers in biopsies from women with palpable DCIS who either developed IBC or who did not. We will concentrate on markers that indicate changes in the stroma since they may explain why the tissue around the lesion becomes stiffer (this is why the DCIS lesion is palpable). This study is important because we now know that changes in the stroma that make it stiffer allow cancer to spread. A cancer will not spread if these changes in the stroma are blocked. Therefore, in a second part of our project, we will test the potential benefit of a new type of drugs recently approved to treat menopausal symptoms. In preliminary studies, this drug seems to inhibit/reverse the changes in the stroma that allow cancer to spread. This could offer new treatment opportunities for women with (palpable) DCIS.

The general methodology: We previously identified markers that, when found in DCIS, correctly predict which women will get IBC years before it forms, and markers that identify women who will not develop future IBC after a pre-cancer diagnosis. Our proposal is based on recent findings that identify changes in the stroma that allow cancer to grow. We use these clues to select new candidate markers and we test them to see how well they predict future IBC. The markers we want to evaluate would reflect changes in the stroma that would be particularly prominent in women diagnosed with palpable DCIS. The markers are found in biopsy tissue samples collected many years ago from women diagnosed with pure DCIS and who had a lumpectomy. Standard practice at that time did not include radiation or hormonal drugs. Since these women agreed to have their tissue studied, to be contacted over time to check on their health, and to learn if they developed an IBC later on, we can see if our targeted markers are present and predict women who developed IBC.

Innovative elements of the project and potential impact: Our first innovation is the selection of a specific group of 15 candidate markers based on published studies and recent findings from our laboratory that have identified a process in the body that makes many changes in the stroma that allow a cancer to spread. We propose to test these markers in breast biopsies from women diagnosed with DCIS by palpation because this process could be active in these women. Our second innovation relates to the approach we will employ to test these 15 markers together. This is like following a conversation by listening to both sides, instead of just one side. Instead of looking at the expression of one marker at a time in tissue sections, we will look at several of them at once. Not only will this make good use of such precious samples, it will also give us information about the conversation between the different cells in the breast tissue and their surrounding environment. Our third innovation is the testing of a drug recently approved by the FDA that can interrupt the conversation that allows the cancer to spread. This drug may represent a new option for women with DCIS to prevent IBC.

Final Report (2019)

We have previously described biological changes (‘markers’) that predict which women diagnosed with Ductal Carcinoma in Situ (DCIS) will develop future invasive breast cancer (IBC). However, our first validated tool (the Abrogated Response to Cellular Stress [ARCS] signature: p16, COX2, and Ki67) is only predictive for ~50% of patients. Our goal is therefore to identify novel markers that will help us better predict progression for women with DCIS. In this project, we are focusing on analyzing breast specimens obtained from women with palpable DCIS in order to identify novel markers that indicate changes in the stroma. These markers may explain why the breast tissue around the DCIS lesion becomes stiffer (why the DCIS lesion is palpable). Notably, patients with such palpable DCIS lesions have a higher risk of progression to IBC. Our recent published studies point to a key role for the protein CD36 in controlling stromal characteristics. We are also investigating the potential benefit of a short-term treatment right after diagnosis of DCIS. The drug we are testing may potentially revert adverse features of the epithelial DCIS lesion and, as importantly, to normalize the surrounding stroma.

To achieve its goal, we proposed to: (1) investigate if and which changes controlled by CD36 are associated with progression of palpated DCIS lesions to IBC by looking at the expression level of novel markers in specimens from women diagnosed with DCIS by palpation and who either subsequently progressed or not to IBC, (2) design a CLIA-approved protocol for automated multiplex staining and software-guided downstream quantitative analysis of multiplex-stained specimens and (3) study if short-term exposure of patients newly diagnosed with DCIS to a candidate drug already approved by the FDA (DuaveeTM) can be beneficial to them by converting a pro-tumorigenic stromal signature into an anti-tumorigenic stromal signature.

As detailed in this last progress report, we have made good progress on all three fronts. Our major accomplishments include: (1) assessing the predictive value of an optimized stromal vasculature signature (CD36/CD31/vimentin) for disease outcome in a pilot cohort of 31 women who progressed to IBC (cases) and 57 who did not (controls); (2) describing our findings in a publication; (3) characterizing in vitro two distinct vasculature populations based on their CD36 or CD31 expression; and (4) accruing the proposed 20 paitent pilot cohort as part of the randomized, double-blinded, placebo-controlled “window of opportunity” pilot trial to test the potential benefit of treatment of women diagnosed with DCIS with DuaveeTM and analyzing this cohort for the vasculature module described above. Our exciting findings that the expression/compartmentalization of our CD36/vasculature biomarker module changes dramatically between disease-free breast tissue, some DCIS lesions and breast cancer and the predictive value of this module for risk stratification strongly suggest the acquisition of a pro-tumorigenic vasculature upon disease progression in some DCIS lesions. These observations strongly support the value of searching for novel predictive “stromal” markers.

Future extension of this work will include: (1) further assessment of the predictive value of this stromal vasculature signature in independent retrospective and prospective cohorts; (2) further assessment of the predictive value of this stromal vasculature signature in relationship with method of detection of the initial DCIS lesions, i.e. mammography versus palpation; (3) pursue our efforts to set up a CLIA-approved platform to further validate the predictive value of this stromal vasculature signature (CD36/CD31/vimentin) as novel risk stratification tool; and (4) further assessment of the effect of DuaveeTM on the stromal landscape surrounding DCIS lesions.