Incorporating Biology Into Breast Cancer Staging: American Joint Committee on Cancer, Eighth Edition, Revisions and Beyond

Male breast cancer: a multicenter study to provide a guide for proper management

INTRODUCTION

To offer an extensive retrospective experience on the management of male breast cancer.

METHODS

A multicenter retrospective observational cohort study was conducted, including male patients diagnosed with breast cancer (invasive or in situ) in 12 Italian breast units from January 1975 to December 2019. Patients aged 18 years or older were assessed for eligibility. Exclusion criteria were metastatic cancer at diagnosis, previous cancer(s), received neoadjuvant treatment, incomplete data on (neo) adjuvant treatment(s), and/or follow-up data. Data on radiological examinations, demographic characteristics, risk factors, histological features, receptor status, treatments, and follow-up were collected.

RESULTS

In a series of 671 male patients with breast cancer assessed for eligibility, 403 (28 in situ and 375 invasive neoplasms) were included in the study. All included patients underwent surgery. The median age at surgery was 63.8 years (IQR 56.1-72.1). In 68% of cases, patients underwent echography, and in 55.1%, a mammography. Most patients were ER and PR positive (63.8%), HER2 negative (80.4%), with high (≥ 20%) Ki67 values (61.3%), and luminal B subtype (51.1%). The 10-year overall survival was 73.6% (95% CI 67.0-79.1) for invasive breast cancer and 90% (95% CI 65.6-97.4) for in situ breast cancer. In patients with invasive breast cancer, at univariable analysis, having a G3 tumor (vs. G1), pT2/3/4 (vs. pT1), pN2/3 (vs. pN0), luminal B subtype with Ki67 ≥ 20% (vs. Luminal A), were significantly associated with a higher risk of death. In multivariable analyses, pT2/3/4 (vs. pT1) remained significantly associated with a higher risk of death (HR 3.14, 95% CI 1.83-5.39), and having a HER2 positive or a triple-negative subtype (vs. Luminal A) was also significantly associated with a higher risk of mortality (HR 4.76, 95% CI 1.26-18.1).

CONCLUSION

Male breast cancer is a rare disease, the better understanding of which is necessary for a more effective diagnostic and therapeutic approach.

Deep neural networks with knockoff features identify nonlinear causal relations and estimate effect sizes in complex biological systems

BACKGROUND

Learning the causal structure helps identify risk factors, disease mechanisms, and candidate therapeutics for complex diseases. However, although complex biological systems are characterized by nonlinear associations, existing bioinformatic methods of causal inference cannot identify the nonlinear relationships and estimate their effect size.

RESULTS

To overcome these limitations, we developed the first computational method that explicitly learns nonlinear causal relations and estimates the effect size using a deep neural network approach coupled with the knockoff framework, named causal directed acyclic graphs using deep learning variable selection (DAG-deepVASE). Using simulation data of diverse scenarios and identifying known and novel causal relations in molecular and clinical data of various diseases, we demonstrated that DAG-deepVASE consistently outperforms existing methods in identifying true and known causal relations. In the analyses, we also illustrate how identifying nonlinear causal relations and estimating their effect size help understand the complex disease pathobiology, which is not possible using other methods.

CONCLUSIONS

With these advantages, the application of DAG-deepVASE can help identify driver genes and therapeutic agents in biomedical studies and clinical trials.

Definition of High-Risk Early Hormone-Positive HER2−Negative Breast Cancer: A Consensus Review

Breast cancer is one of the major causes of cancer-related morbidity and mortality in women worldwide. During the past three decades, several improvements in the adjuvant treatment of hormone receptor-positive/HER2−negative breast cancer have been achieved with the introduction of optimized adjuvant chemotherapy and endocrine treatment. However, estimating the risk of relapse of breast cancer on an individual basis is still challenging. The IRIDE (hIGh Risk DEfinition in breast cancer) working group was established with the aim of reviewing evidence from the literature to synthesize the current relevant features that predict hormone-positive/HER2−negative early breast cancer relapse. A panel of experts in breast cancer was involved in identifying clinical, pathological, morphological, and genetic factors. A RAND consensus method was used to define the relevance of each risk factor. Among the 21 features included, 12 were considered relevant risk factors for relapse. For each of these, we provided a consensus statement and relevant comments on the supporting scientific evidence. This work may guide clinicians in the practical management of hormone-positive/HER2−negative early breast cancers.

A Novel Staging System for De Novo Metastatic Breast Cancer Refines Prognostic Estimates

OBJECTIVE

We aim to identify prognostic groups within a de novo metastatic cohort, incorporating both anatomic and biologic factors.

BACKGROUND

Staging for breast cancer now includes anatomic and biologic factors, although the guidelines for stage IV disease do not account for how these factors may influence outcomes.

METHODS

Adults with de novo metastatic breast cancer were selected from the National Cancer DataBase (2010-2013). Recursive partitioning analysis was used to group patients with similar overall survival (OS) based on clinical T/N stage, tumor grade, ER, PR, HER2, number of metastatic sites, and presence of bone-only metastases. Categories were created by amalgamating homogeneous groups based on 3-year OS rates (stage IVA: >50%, stage IVB: 30%-50%, stage IVC: <30%).

RESULTS

16,187 patients were identified; median follow-up was 32 months. 65.2% had 1 site of distant metastasis, and 42.9% had bone-only metastases. Recursive partitioning analysis identified the number of metastatic sites (1 vs >1) as the first stratification point, and ER status as the second stratification point for both resulting groups. Additional divisions were made based on HER2 status, PR status, cT stage, tumor grade, and presence of bone-only metastases. After bootstrapping, significant differences in 3-year OS were noted between the 3 groups [stage IVB vs IVA: HR 1.58 (95% confidence interval 1.50-1.67), stage IVC vs IVA: HR 3.54 (95% confidence interval 3.33-3.77)].

CONCLUSIONS

Both anatomic and biologic factors yielded reliable and reproducible prognostic estimates among patients with metastatic disease. These findings support formal stratification of de novo stage IV breast cancer into 3 distinct prognosis groups.

OUP accepted manuscript

Background

Methods

Results

Conclusion

Looking for more reliable biomarkers in breast cancer: Comparison between routine methods and RT-qPCR

PURPOSE

Decades of quality control efforts have raised the standards of immunohistochemistry (IHC), the principle method used for biomarker testing in breast cancer; however, computational pathology and reverse transcription quantitative PCR (RT-qPCR) may also hold promise for additional substantial improvements.

METHODS

Herein, we investigated discrepancies in the assessment of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and marker of proliferation Ki67 comparing routinely obtained IHC (and FISH) data (ORI) with the results of manual (REV) and semi-automated (DIA) re-evaluation of the original IHC slides and then with RNA expression data from the same tissue block using the MammaTyper® (MT) gene expression assay.

RESULTS

Correlation for ER and PR was high between ORI IHC and the other three study methods (REV, DIA and RT-qPCR). For HER2, 10 out of 96 discrepant cases can be detected between ORI and REV that involved at least one call in the equivocal category (except for one case). For Ki67, 22 (29.1%) cases were categorized differently by either REV alone (n = 17), DIA alone (n = 15) or both (n = 10) and 28 cases (29.2%) for RT-qPCR. Most of the discrepant Ki67 cases changed from low to high between the original and following assessment and belonged to the intermediate Ki67 expression range (between 9 and 30%).

CONCLUSIONS

Determination of the breast cancer biomarkers ER, PR, HER2 and Ki67 at the mRNA level shows high degree of correlation with IHC and compares well with correlations between original with subsequent independent manual or semi-automated IHC assessments. The use of methods with wider dynamic range and higher reproducibility such as RT-qPCR may offer more precise assessment of endocrine responsiveness, improve Ki67 standardization and help resolve HER2 cases that remain equivocal or ambiguous by IHC/FISH. In summary, our findings seem to configure RT-qPCR as a complementary method to be used in cases of either equivocal results or presenting, at the traditional determination assays, biomarkers expressions close to the cut-off values.

Comparison of Breast Cancer Staging Systems After Neoadjuvant Chemotherapy

BACKGROUND

No consensus exists for optimal staging following neoadjuvant chemotherapy (NAC). We compared the performance of the American Joint Committee on Cancer (AJCC) pathologic prognostic staging system, Residual Cancer Burden (RCB) Index, and the Neo-Bioscore in breast cancer patients after NAC.

METHODS

Patients with stage I-III breast cancer who received NAC at Dana-Farber Cancer Institute from 2004 to 2014 were identified. Kaplan-Meier curves were used to estimate disease-free survival (DFS) and overall survival (OS), and model fits were compared by receiver operator characteristic (ROC) curve using the c-statistic and DeLong's test.

RESULTS

Overall, 802 patients with a median age of 48 years received NAC. Most patients presented with cT2 (n = 470, 58.6%) and cN1 (n = 422, 52.6%) disease. The subtype was estrogen receptor (ER)- and/or progesterone receptor (PR)-positive/human epidermal growth factor receptor 2 (HER2)-negative in 296 (36.9%) patients, HER2-positive in 261 (32.5%) patients, and triple-negative in 245 (30.5%) patients. Median follow-up was 79.5 months. There were 174 recurrences (30 local, 25 regional, 145 distant), with 676 (76.8%) patients alive at last follow-up. AJCC pathologic prognostic staging and RCB had better discrimination for estimated 7-year DFS and OS compared with the Neo-Bioscore. The ROC c-statistics for DFS model fit were similar for AJCC pathologic prognostic stage (0.72) and RCB (0.71, p = non-significant); both had improved model fit versus the Neo-Bioscore (0.65, p < 0.01). The c-statistics for OS were 0.74, 0.71, and 0.70 for AJCC pathologic prognostic stage, RCB, and Neo-Bioscore, respectively (p = non-significant).

CONCLUSIONS

These results validate the ability of these staging systems to stratify survival outcomes in NAC patients, with best discrimination achieved using AJCC pathologic prognostic stage or RCB.

Co-expression based cancer staging and application

A novel method is developed for predicting the stage of a cancer tissue based on the consistency level between the co-expression patterns in the given sample and samples in a specific stage. The basis for the prediction method is that cancer samples of the same stage share common functionalities as reflected by the co-expression patterns, which are distinct from samples in the other stages. Test results reveal that our prediction results are as good or potentially better than manually annotated stages by cancer pathologists. This new co-expression-based capability enables us to study how functionalities of cancer samples change as they evolve from early to the advanced stage. New and exciting results are discovered through such functional analyses, which offer new insights about what functions tend to be lost at what stage compared to the control tissues and similarly what new functions emerge as a cancer advances. To the best of our knowledge, this new capability represents the first computational method for accurately staging a cancer sample. The R source code used in this study is available at GitHub (https://github.com/yxchspring/CECS).

Validation of the AJCC prognostic stage for HER2-positive breast cancer in the ShortHER trial

BACKGROUND

The 8th edition of the American Joint Committee on Cancer (AJCC) staging has introduced prognostic stage based on anatomic stage combined with biologic factors. We aimed to validate the prognostic stage in HER2-positive breast cancer patients enrolled in the ShortHER trial.

METHODS

The ShortHER trial randomized 1253 HER2-positive patients to 9 weeks or 1 year of adjuvant trastuzumab combined with chemotherapy. Patients were classified according to the anatomic and the prognostic stage. Distant disease-free survival (DDFS) was calculated from randomization to distant relapse or death.

RESULTS

A total of 1244 patients were included. Compared to anatomic stage, the prognostic stage downstaged 41.6% (n = 517) of patients to a more favorable stage category. Five-year DDFS based on anatomic stage was as follows: IA 96.6%, IB 94.1%, IIA 92.4%, IIB 87.3%, IIIA 81.3%, IIIC 70.5% (P < 0.001). Five-year DDFS according to prognostic stage was as follows: IA 95.7%, IB 91.4%, IIA 86.9%, IIB 85.0%, IIIA 77.6%, IIIC 67.7% (P < 0.001). The C index was similar (0.69209 and 0.69249, P = 0.975). Within anatomic stage I, the outcome was similar for patients treated with 9 weeks or 1 year trastuzumab (5-year DDFS 96.2% and 96.6%, P = 0.856). Within prognostic stage I, the outcome was numerically worse for patients treated with 9 weeks trastuzumab (5-year DDFS 93.7% and 96.3%, P = 0.080).

CONCLUSIONS

The prognostic stage downstaged 41.6% of patients, while maintaining a similar prognostic performance as the anatomic stage. The prognostic stage is valuable in counseling patients and may serve as reference for a clinical trial design. Our data do not support prognostic stage as guidance to de-escalate treatment.

TRIAL REGISTRATION

EUDRACT number: 2007-004326-25; NCI ClinicalTrials.gov number: NCT00629278.

Insights on CTC Biology and Clinical Impact Emerging from Advances in Capture Technology

Circulating tumor cells (CTCs) and circulating tumor microemboli (CTM) have been shown to correlate negatively with patient survival. Actual CTC counts before and after treatment can be used to aid in the prognosis of patient outcomes. The presence of circulating tumor materials (CTMat) can advertise the presence of metastasis before clinical presentation, enabling the early detection of relapse. Importantly, emerging evidence is indicating that cancer treatments can actually increase the incidence of CTCs and metastasis in pre-clinical models. Subsequently, the study of CTCs, their biology and function are of vital importance. Emerging technologies for the capture of CTC/CTMs and CTMat are elucidating vitally important biological and functional information that can lead to important alterations in how therapies are administered. This paves the way for the development of a "liquid biopsy" where treatment decisions can be informed by information gleaned from tumor cells and tumor cell debris in the blood.