Abstract P1-06-02: Mismatch repair protein loss in breast cancer: Clinicopathological associations in a large British Columbia cohort
Bibliographic record
Abstract
Abstract Background: Alterations to mismatched repair (MMR) pathways are a known cause of cancer (particularly colorectal and endometrial). Recently, the FDA approved pembrolizumab for use in MMR-deficient (MMRD) cancers of any type, and the diagnosis can be made by immunohistochemistry (IHC) or genomic methods. In breast cancer, mutational process analyses indicate MMRD occurs in about 2% of breast cancer (Cancer Res; 77; 4755-62, 2017) and recent functional studies have shown associations with resistance to endocrine therapy and sensitivity to CDK4/6 inhibitors (Cancer Discov; 7; 1168-83, 2017). To date, insufficient cases have been assembled to power meaningful associative or survival studies. Herein, the strong correlation between IHC-determined loss of MLH1, PMS2, MSH2 or MSH6 and genomic evidence allowed the assessment of MMRD on a large tissue microarray (TMA) series linked to detailed biomarkers and long-term outcome data. Methods: IHC markers MLH1, PMS2, MSH2 and MSH6 were optimized on the Ventana automated stainer for application to breast cancer TMAs. The patient cohort consists of females from British Columbia diagnosed with primary invasive breast carcinoma in 1986-1992, referred to the British Columbia Cancer Agency for treatment and follow-up. TMA blocks were sectioned and stained. Slides were scored by a pathologist and only nuclear positivity was evaluated positive. Loss of nuclear positivity for any one of the four tested marker defined MMRD. Clinicopathological associations were tested by Chi-square, and survival by Kaplan-Meier plot with log rank test. Result: 1635 cases were interpretable for all MMR markers. 31 cases (1.9%) met criteria for MMRD. 6 cases had paired losses (4 MLH1-PMS2 loss, 2 MSH2-MSH6 loss) and the remaining 25 cases had singular MMR loss (11 PMS2 loss, 10 MLH1 loss, 3 MSH6 loss, 1 MSH2 loss). Deficiency of the the MutL complex (MLH1/PMS2) predominated over the MutS complex (MSH2/MSH6). Among the demographic and pathological variables assessed – age, grade, tumour size, lymphovascular invasion, nodal and menstrual status – high grade is associated with MMRD (p=0.014). In terms of biomarker, MMRD is significantly associated with PR negativity (p=0.003) and PD-L1 expression (p=0.049), but not with ER, Her2, Ki67, or basal breast cancer IHC markers, nor does MMRD significantly correlate with any of the established major intrinsic subtypes of breast cancer. Tumor infiltrating lymphocyte (TIL) counts are higher in MMRD cases (p=0.009). Although statistically not significant (small numbers), Kaplan-Meier plots of survival analysis demonstrated a trend for MMR loss to be associated with decreased breast cancer disease-specific and overall survival. Conclusion: This large series assessed by IHC corroborates findings from smaller genomic series that MMRD is present in about 2% of breast cancers. MMRD tumors are more likely to be high grade, low PR and immunologically active (higher PD-L1 expression and TIL counts). MMR deficiency is present across all major molecular subtypes (luminal, HER2, basal). Given the efficacy of PD1/PDL1 targeting agents in MMR deficient tumors of other types, evidence for the activity of these agents in MMR deficient breast cancers should be actively sought. Citation Format: Cheng AS, Leung SC, Gao D, Anurag M, Nielsen T, Ellis MJ. Mismatch repair protein loss in breast cancer: Clinicopathological associations in a large British Columbia cohort [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P1-06-02.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".