Validation of the OncoMasTR Risk Score in Estrogen Receptor–Positive/HER2-Negative Patients: A TransATAC study

Clinically relevant gene signatures provide independent prognostic information in older breast cancer patients

BACKGROUND

The clinical utility of gene signatures in older breast cancer patients remains unclear. We aimed to determine signature prognostic capacity in this patient subgroup.

METHODS

Research versions of the genomic grade index (GGI), 70-gene, recurrence score (RS), cell cycle score (CCS), PAM50 risk-of-recurrence proliferation (ROR-P), and PAM50 signatures were applied to 39 breast cancer datasets (N = 9583). After filtering on age ≥ 70 years, and the presence of estrogen receptor (ER) and survival data, 871 patients remained. Signature prognostic capacity was tested in all (n = 871), ER-positive/lymph node-positive (ER + /LN + , n = 335) and ER-positive/lymph node-negative (ER + /LN-, n = 374) patients using Kaplan-Meier and multivariable Cox-proportional hazard (PH) modelling.

RESULTS

All signatures were statistically significant in Kaplan-Meier analysis of all patients (Log-rank P < 0.001). This significance remained in multivariable analysis (Cox-PH, P ≤ 0.05). In ER + /LN + patients all signatures except PAM50 were significant in Kaplan-Meier analysis (Log-rank P ≤ 0.05) and remained so in multivariable analysis (Cox-PH, P ≤ 0.05). In ER + /LN- patients all except RS were significant in Kaplan-Meier analysis (Log-rank P ≤ 0.05) but only the 70-gene, CCS, ROR-P, and PAM50 signatures remained so in multivariable analysis (Cox-PH, P ≤ 0.05).

CONCLUSIONS

We found that gene signatures provide prognostic information in survival analyses of all, ER + /LN + and ER + /LN- older (≥ 70 years) breast cancer patients, suggesting a potential role in aiding treatment decisions in older patients.

Applying a GAN-based classifier to improve transcriptome-based prognostication in breast cancer

Established prognostic tests based on limited numbers of transcripts can identify high-risk breast cancer patients yet are approved only for individuals presenting with specific clinical features or disease characteristics. Deep learning algorithms could hold potential for stratifying patient cohorts based on full transcriptome data, yet the development of robust classifiers is hampered by the number of variables in omics datasets typically far exceeding the number of patients. To overcome this hurdle, we propose a classifier based on a data augmentation pipeline consisting of a Wasserstein generative adversarial network (GAN) with gradient penalty and an embedded auxiliary classifier to obtain a trained GAN discriminator (T-GAN-D). Applied to 1244 patients of the METABRIC breast cancer cohort, this classifier outperformed established breast cancer biomarkers in separating low- from high-risk patients (disease specific death, progression or relapse within 10 years from initial diagnosis). Importantly, the T-GAN-D also performed across independent, merged transcriptome datasets (METABRIC and TCGA-BRCA cohorts), and merging data improved overall patient stratification. In conclusion, the reiterative GAN-based training process allowed generating a robust classifier capable of stratifying low- vs high-risk patients based on full transcriptome data and across independent and heterogeneous breast cancer cohorts.

A radiomic model to classify response to neoadjuvant chemotherapy in breast cancer

BACKGROUND

Medical image analysis has evolved to facilitate the development of methods for high-throughput extraction of quantitative features that can potentially contribute to the diagnostic and treatment paradigm of cancer. There is a need for further improvement in the accuracy of predictive markers of response to neo-adjuvant chemotherapy (NAC). The aim of this study was to develop a radiomic classifier to enhance current approaches to predicting the response to NAC breast cancer.

METHODS

Data on patients treated for breast cancer with NAC prior to surgery who had a pre-NAC dynamic contrast enhanced breast MRI were included. Response to NAC was assessed using the Miller-Payne system on the excised tumor. Tumor segmentation was carried out manually under the supervision of a consultant breast radiologist. Features were selected using least absolute shrinkage selection operator regression. A support vector machine learning model was used to classify response to NAC.

RESULTS

74 patients were included. Patients were classified as having a poor response to NAC (reduction in cellularity < 90%, n = 44) and an excellent response (> 90% reduction in cellularity, n = 30). 4 radiomics features (discretized kurtosis, NGDLM contrast, GLZLM_SZE and GLZLM_ZP) were identified as pertinent predictors of response to NAC. A SVM model using these features stratified patients into poor and excellent response groups producing an AUC of 0.75. Addition of estrogen receptor status improved the accuracy of the model with an AUC of 0.811.

CONCLUSION

This study identified a radiomic classifier incorporating 4 radiomics features to augment subtype based classification of response to NAC in breast cancer.

The OncoMasTR test predicts distant recurrence in estrogen receptor-positive, HER2-negative early-stage breast cancer: A validation study in ABCSG Trial 8

PURPOSE

To validate the clinical performance of the OncoMasTR Risk Score in the biomarker cohort of ABCSG Trial 8.

EXPERIMENTAL DESIGN

We evaluated the OncoMasTR test in 1200 formalin-fixed, paraffin-embedded surgical specimens from postmenopausal women with estrogen receptor (ER)-positive, human epidermal growth factor receptor 2 (HER2)-negative primary breast cancer with 0-3 involved lymph nodes in the prospective, randomized ABCSG Trial 8. Time to distant recurrence (DR) was analyzed by Cox models.

RESULTS

The OncoMasTR Risk Score categorized 850 of 1087 (78.2%) evaluable patients as "low risk". At 10 years, the DR rate for patients in the low-risk group was 5.8% versus 21.1% for patients in the high-risk group (P<0.0001, absolute risk reduction 15.3%). The OncoMasTR Risk Score was highly prognostic for prediction of DR in years 0-10 in all patients (hazard ratio (HR) 1.91, 95% confidence interval (CI) 1.62 to 2.26, P<0.0001; C-index 0.73), in node-negative patients (HR 1.79, 95% CI 1.43 to 2.24, P<0.0001; C-index 0.72), and in patients with 1-3 involved lymph nodes (HR 1.93, 95% CI 1.44 to 2.58, P<0.0001; C-index 0.71). The OncoMasTR Risk Score provided significant additional prognostic information beyond clinical parameters, Ki67, Nottingham Prognostic Index, and Clinical Treatment Score.

CONCLUSIONS

OncoMasTR Risk Score is highly prognostic for DR in postmenopausal women with ER-positive, HER2-negative primary breast cancer with 0-3 involved lymph nodes. In combination with prior validation studies, this fully independent validation in ABCSG Trial 8 provides level 1B evidence for the prognostic capability of the OncoMasTR Risk Score.

Future of biomarker evaluation in the realm of artificial intelligence algorithms: application in improved therapeutic stratification of patients with breast and prostate cancer

Clinical workflows in oncology depend on predictive and prognostic biomarkers. However, the growing number of complex biomarkers contributes to costly and delayed decision-making in routine oncology care and treatment. As cancer is expected to rank as the leading cause of death and the single most important barrier to increasing life expectancy in the 21st century, there is a major emphasis on precision medicine, particularly individualisation of treatment through better prediction of patient outcome. Over the past few years, both surgical and pathology specialties have suffered cutbacks and a low uptake of pathology specialists means a solution is required to enable high-throughput screening and personalised treatment in this area to alleviate bottlenecks. Digital imaging in pathology has undergone an exponential period of growth. Deep-learning (DL) platforms for hematoxylin and eosin (H&E) image analysis, with preliminary artificial intelligence (AI)-based grading capabilities of specimens, can evaluate image characteristics which may not be visually apparent to a pathologist and offer new possibilities for better modelling of disease appearance and possibly improve the prediction of disease stage and patient outcome. Although digital pathology and AI are still emerging areas, they are the critical components for advancing personalised medicine. Integration of transcriptomic analysis, clinical information and AI-based image analysis is yet an uncultivated field by which healthcare professionals can make improved treatment decisions in cancer. This short review describes the potential application of integrative AI in offering better detection, quantification, classification, prognosis and prediction of breast and prostate cancer and also highlights the utilisation of machine learning systems in biomarker evaluation.

Prognostic value of the 6-gene OncoMasTR test in hormone receptor-positive HER2-negative early-stage breast cancer: Comparative analysis with standard clinicopathological factors

AIM

The aim of the study was to assess the prognostic performance of a 6-gene molecular score (OncoMasTR Molecular Score [OMm]) and a composite risk score (OncoMasTR Risk Score [OM]) and to conduct a within-patient comparison against four routinely used molecular and clinicopathological risk assessment tools: Oncotype DX Recurrence Score, Ki67, Nottingham Prognostic Index and Clinical Risk Category, based on the modified Adjuvant! Online definition and three risk factors: patient age, tumour size and grade.

METHODS

Biospecimens and clinicopathological information for 404 Irish women also previously enrolled in the Trial Assigning Individualized Options for Treatment [Rx] were provided by 11 participating hospitals, as the primary objective of an independent translational study. Gene expression measured via RT-qPCR was used to calculate OMm and OM. The prognostic value for distant recurrence-free survival (DRFS) and invasive disease-free survival (IDFS) was assessed using Cox proportional hazards models and Kaplan-Meier analysis. All statistical tests were two-sided ones.

RESULTS

OMm and OM (both with likelihood ratio statistic [LRS] P < 0.001; C indexes = 0.84 and 0.85, respectively) were more prognostic for DRFS and provided significant additional prognostic information to all other assessment tools/factors assessed (all LRS P ≤ 0.002). In addition, the OM correctly classified more patients with distant recurrences (DRs) into the high-risk category than other risk classification tools. Similar results were observed for IDFS.

DISCUSSION

Both OncoMasTR scores were significantly prognostic for DRFS and IDFS and provided additional prognostic information to the molecular and clinicopathological risk factors/tools assessed. OM was also the most accurate risk classification tool for identifying DR. A concise 6-gene signature with superior risk stratification was shown to increase prognosis reliability, which may help clinicians optimise treatment decisions.

Multi-Gene Prognostic Signatures and Prediction of Pathological Complete Response to Neoadjuvant Chemotherapy in ER-positive, HER2-negative Breast Cancer Patients

Determining which patients with early-stage breast cancer should receive chemotherapy is an important clinical issue. Chemotherapy has several adverse side effects, impacting on quality of life, along with significant economic consequences. There are a number of multi-gene prognostic signatures for breast cancer recurrence but there is less evidence that these prognostic signatures are predictive of therapy benefit. Biomarkers that can predict patient response to chemotherapy can help avoid ineffective over-treatment. The aim of this work was to assess if the OncoMasTR prognostic signature can predict pathological complete response (pCR) to neoadjuvant chemotherapy, and to compare its predictive value with other prognostic signatures: EndoPredict, Oncotype DX and Tumor Infiltrating Leukocytes. Gene expression datasets from ER-positive, HER2-negative breast cancer patients that had pre-treatment biopsies, received neoadjuvant chemotherapy and an assessment of pCR were obtained from the Gene Expression Omnibus repository. A total of 813 patients with 66 pCR events were included in the analysis. OncoMasTR, EndoPredict, Oncotype DX and Tumor Infiltrating Leukocytes numeric risk scores were approximated by applying the gene coefficients to the corresponding mean probe expression values. OncoMasTR, EndoPredict and Oncotype DX prognostic scores were moderately well correlated according to the Pearson's correlation coefficient. Association with pCR was estimated using logistic regression. The odds ratio for a 1 standard deviation increase in risk score, adjusted for cohort, were similar in magnitude for all four signatures. Additionally, the four signatures were significant predictors of pCR. OncoMasTR added significant predictive value to EndoPredict, Oncotype DX and Tumor Infiltrating Leukocytes signatures as determined by bivariable and trivariable analysis. In this in silico analysis, OncoMasTR, EndoPredict, Oncotype DX, and Tumor Infiltrating Leukocytes were significantly predictive of pCR to neoadjuvant chemotherapy in ER-positive and HER2-negative breast cancer patients.

Breast Cancer: A Molecularly Heterogenous Disease Needing Subtype-Specific Treatments

Breast cancer is the most commonly occurring cancer in women. There were over two-million new cases in world in 2018. It is the second leading cause of death from cancer in western countries. At the molecular level, breast cancer is a heterogeneous disease, which is characterized by high genomic instability evidenced by somatic gene mutations, copy number alterations, and chromosome structural rearrangements. The genomic instability is caused by defects in DNA damage repair, transcription, DNA replication, telomere maintenance and mitotic chromosome segregation. According to molecular features, breast cancers are subdivided in subtypes, according to activation of hormone receptors (estrogen receptor and progesterone receptor), of human epidermal growth factors receptor 2 (HER2), and or BRCA mutations. In-depth analyses of the molecular features of primary and metastatic breast cancer have shown the great heterogeneity of genetic alterations and their clonal evolution during disease development. These studies have contributed to identify a repertoire of numerous disease-causing genes that are altered through different mutational processes. While early-stage breast cancer is a curable disease in about 70% of patients, advanced breast cancer is largely incurable. However, molecular studies have contributed to develop new therapeutic approaches targeting HER2, CDK4/6, PI3K, or involving poly(ADP-ribose) polymerase inhibitors for BRCA mutation carriers and immunotherapy.