Abstract P2-08-52: A predictive model for distant metastasis in breast cancer patients using machine learning

Introduction

Tumor metastasis is a major clinical challenge accounting for the vast majority of cancer related deaths.In previous studies, prediction of distant metastasis was based on subtypes,clinical status and sometimes gene expression were used however clinical application was difficult.

In this study, we develop the easy to use prediction tool for distant metastasis using clinical characteristics and gene profiles which came from CancerSCANTM, Next Generation Sequencing based targeted-sequencing platform designed at Samsung Medical Center(SMC).

Methods

We performed a retrospective chart review of 326 breast cancer patients who underwent surgery and CancerSCAN TM between Jan 2001 and Dec 2014 at SMC. Median follow up period was 83 months (Range 1˜190). Cancer scanTM cover 381 genes but 27 genes and 34 occasions (loss of function, mutation or copy number variation) were selected for analysis through gradient boosting and Wilcoxon Signed rank test. Azure Machine Learning is a cloud service that enables the execution of machine learning processes.This was accomplished using the steps of (1) edit the data, (2) split the data, (3) train the model, (4) score the model, and (5) evaluate the model. We split the modeling data into training and testing sets using a randomized 50–50 split. Two-class Decision Forest method was used. After deploying the Azure ML predictive model as a web service, we used a Representational State Transfer application programming interface to send data and obtained predictions in real-time.

Results

No distant metastasis group and distant metastasis group consisted of 267 and 59 patients, respectively. HR-/HER2+ and 50 years old and over patients were higher in metastasis group (p-value = 0.003 and p-value = 0.000). Nuclear grade 3 and N2,3 were higher in metastasis group (p-value = 0.010 and p-value = 0.000, p-value = 0.001 respectively). Stage III was also higher in metastasis group (p-value = 0.000). Among 59 patients with distantmetastasis, multiple sites metastasis was 21 cases (35.6%) and then lung metastasis was 19 cases (32.2%). In the 21 cases of multiple sites metastasis, triple sites was 6 cases (28.6%) and double sites was 15 cases (71.4%). PIK3CA mutation was the most frequent gene variation in all patients (34.5% of no metastasis group and 27.1% of metastasis group) but there was no difference between two groups(p-value = 0.278). BRCA 1 loss of function and BRCA2 loss of function were more frequent in metastasis group than no metastasis group(p-value = 0.033 and p-value = 0.024, respectively) but total counts was too small. We assessed the area under the curve (AUC) of the receiver operating characteristic (ROC) curve for predictive value. The AUC of ROC curve was 1.000 and also accuracy, precision, recall were 1.000. In addition, we conducted internal validation using 83 patients during 2015. When we applied a 0.5 threshold value with our predictive model, true negative was 81 and true positive was 2 among 83 patients. Finally, the accuracy of validation was 1.000.

Conclusion

Our predicted model could represent a useful and easy-to-access tool for the selection of patients with distant metastasis. After additional evaluation with large data and external validation, worldwide use of our model could be expected.

Citation Format: Kim I, Choi HJ, Ryu JM, Lee SK, Yu JH, Kim SW, Nam SJ, Seo SW, Lee JE. A predictive model for distant metastasis in breast cancer patients using machine learning [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 P2-08-52.

Abstract P1-15-18: Not presented

This abstract was not presented at the conference.

Citation Format: Choi HJ, Kim SW, Ryu JM, Kim I, Nam SJ, Yu J, Lee SK, Lee JE. Not presented [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 P1-15-18.

Abstract P2-08-29: The impact of time interval between diagnosis and surgery in each type and stage of breast cancer

Background: There are many factors that might contribute to the delay of surgery in patients with breast cancer. Previous studies investigate the influence of delay of surgery, but they reported inconsistent results. The purpose of this study was to evaluate the impact of time of surgery on prognosis of breast cancer.

Methods: We performed a retrospective review of the patients with breast cancer, who received surgery between 1992 and 2009, by using data from Korea Breast Cancer Society Registry. Kaplan-Meier survival analysis and Cox regression model were used to evaluate the impact of time to surgery in breast cancer and subgroup analyses were performed for each disease stage and molecular subtype.

Result: A total 14727 patients were included for analysis. Delay of surgery more than 31 days was associated with worse survival for breast cancer [hazard ratio (HR) = 2.16; 95% confidence interval (CI), 1.936-2.408, p<0.001]. Subgroup analyses revealed that over 31 days of surgical delay were significantly associated with worse survival in hormone receptor positive and HER-2 negative (p<0.001), hormone receptor positive and HER-2 positive (p<0.001), hormone receptor negative and HER-2 positive (p<0.001), triple negative (p<0.001) and stage II, III breast cancer patients (p<0.001).

Conclusion: Surgical delay of more than 31 days were independent risk factors for worse outcome of breast cancer in each molecular subtype and breast cancer group except stage 0 and I. Although preoperative evaluation is required, surgical delay should be shortened to enhance survival of breast cancer, especially in patients with tumor size more than 2cm or presence of lymph node metastasis.

Citation Format: Kim J-M, Choi HJ, Kim I, Ryu JM, Yu J, Lee JE, Kim SW, Nam SJ, Lee SK. The impact of time interval between diagnosis and surgery in each type and stage of breast cancer [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 P2-08-29.

Abstract P2-09-03: Updated efficacy of first or second-line pembrolizumab (pembro) plus capecitabine (cape) in metastatic triple negative breast cancer (mTNBC) and correlations with baseline lymphocyte and naïve CD4+ T-cell count

Background: In mTNBC, anti-PD-1/L1 monotherapy is most effective when administered early in the course of disease, with recent trials demonstrating overall response rates (ORR) of 23-26% in the first-line setting and 5-6% in later lines. This may reflect iatrogenic lymphopenia from preceding cytotoxic chemotherapy. Furthermore, curative-intent chemotherapy is associated with prolonged suppression of naïve CD4+ cells, a T-cell subset that may play a critical role in the generation of de novo anti-tumor immune responses. We present the final clinical results of a pilot study evaluating the safety and efficacy of combining pembrolizumab plus standard-of-care capecitabine in the first/second-line mTNBC setting. We also explore potential associations between clinical benefit and lymphopenia, preceding chemotherapy, and absolute naïve CD4+ counts.

Methods: In a pilot study, we evaluated the tolerability and preliminary efficacy of concurrent pembro (200mg IV q21 day) plus investigator-selected 1st/2nd line paclitaxel (80mg/m2 IV weekly) or oral cape (2,000mg BID, weekly 1 on/1 off). The primary endpoint was tolerability, defined as the proportion of subjects receiving >6 weeks concurrent therapy without dose discontinuation with toxicities reported per CTCAE v4.0. The secondary endpoint was 12-week objective response rate (ORR) by RECIST1.1. Exploratory endpoints included peripheral blood cell enumeration by real-time flow cytometry and routine clinical laboratory. Naïve CD4+ cells were defined as CD45+ CD3+ TCRab+ CD4+ CD45RA+ CCR7+. Here, we report the results of the pilot phase of the cape cohort (NCT02734290).

Results: Twelve of 14 subjects were treated in the first-line setting. All subjects (14/14, 100%) tolerated cape+pembro for >6 weeks, with toxicities consistent with monotherapy cape experience (diarrhea: grade I-II 50%, grade III 7%; hand-foot: grade I-II 71%) that improved with dose-reduction as needed. At 12 weeks, the ORR was 6/14 (42.9%), and the clinical benefit rate (ORR + stable disease) was 8/14 (57.1%). Depressed absolute lymphocyte count at baseline (ALC<1.0/uL: 33% CBR; ALC≥1.0/uL: 75% CBR) and recent exposure to cytotoxic chemotherapy (<6 months: 33% CBR; >6 months: 75% CBR) were associated with reduced clinical benefit. By flow cytometry, subjects experiencing clinical benefit had higher baseline absolute naïve CD4+ counts (average 283 cells/uL v. 93 cells/uL, p=.069).

Conclusions: This study met the primary endpoint of safety for cape plus pembro in mTNBC, with encouraging clinical activity. These data are supportive of further studies evaluating combination chemotherapy plus anti-PD-1/L1 mTNBC. We observed greater clinical benefit in subjects with non-suppressed ALC, less exposure to recent chemo, and higher baseline naïve CD4+ counts, suggesting that iatrogenic immunosuppression can impair response to immune checkpoint therapy in mTNBC. These findings should be confirmed in ongoing randomized trials of immune checkpoint +/- chemotherapy in mTNBC, and should be considered in the design of future clinical trials.

Citation Format: Page DB, Pucilowska J, Bennetts L, Kim I, Sanchez K, Martel M, Conlin A, Moxon N, Mellinger S, Acheson A, Kemmer K, Mitri Z, Vuky J, Ahn J, Abaya C, Manigault T, Basho R, Urba WJ, McArthur HL. Updated efficacy of first or second-line pembrolizumab (pembro) plus capecitabine (cape) in metastatic triple negative breast cancer (mTNBC) and correlations with baseline lymphocyte and naïve CD4+ T-cell count [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 P2-09-03.

New onset acute pulmonary edema after methylergonovine given during cesarean delivery of a patient with undiagnosed Raynaud's disease

Raynaud's disease is a medical condition in which arterial spasm causes episodes of reduced blood flow, in the setting of certain triggers such as cold weather. Patients with this condition are at risk of adverse reactions if they receive medications with vasoactive properties. Methylergonovine maleate is one drug used during cesarean delivery to treat postpartum hemorrhage due to uterine atony. By acting directly on uterine and vascular smooth muscle, it produces cardiovascular responses such as coronary vasospasm, myocardial infarction, and even cardiac arrest. However, pulmonary events have rarely been reported. We report our anesthetic management of a 36-year-old patient, with undiagnosed Raynaud's disease and undergoing cesarean delivery, who experienced new onset acute pulmonary edema after methylergonovine administration to manage postpartum hemorrhage.