Machine learning with multiparametric breast MRI for prediction of Ki-67 and histologic grade in early-stage luminal breast cancer

146 Background: Oncotype DX is an RT-PCR based assay that calculates a recurrence score (RS) to predict chemotherapy benefit in patients with breast cancer. Some authors have proposed using Ki-67, receptor status, Nottingham grade, and tumor size as histo-pathologic variables that could potentially substitute for the RS. Comprehensive and larger reviews are still lacking in this filed. To our knowledge this is the largest retrospective review of cases in a single academic institution. Methods: We retrospectively reviewed and reported baseline patient demographic characteristics (including age, race and gender) and routine pathological features such as histologic grade, Ki-67, tumor size, and histologic type. All patients included in this study had an Oncotype Dx ordered between 2007 and 2014. Results: Our analysis included 252 patients of which 248 were females and 4 were males, median age was 56, median tumor size of 2.1 cm. The majority of cases were grade 2 with 49.4%, followed by grade 3 with 28.1% and grade one represented 22.5% of the sample. Ki-67 ranged from 2-98 %. Three variables correlated significantly with Oncotype Dx: tumor size (p = 0.0115), Ki-67 (p = 1.509e-09) and grade (p = 0.0307). Using these three variables together, the percentage of variance on the Oncotype Dx score was 0.35 (R2- coefficient of determination). Conclusions: While tumor size, grade and Ki-67 individually correlate significantly with Oncotype Dx score, taken together, they cannot reliably predict the RS. Also, the capability of these three factors to predict the RS decreases for tumors with high RS. Therefore, we conclude the RS can provide additional valuable information and should not be replaced by analysis of routine histologic variables alone.

Download Full-text

Prognostic relevance at 5 years of the early monitoring of neoadjuvant chemotherapy using FDG PET in luminal HER2-negative breast cancer.

Journal of Clinical Oncology ◽

10.1200/jco.2013.31.15_suppl.11007 ◽

2013 ◽

Vol 31 (15_suppl) ◽

pp. 11007-11007

Author(s):

Olivier Humbert ◽

Alina Berriolo-Riedinger ◽

Alexandre Cochet ◽

Mélanie Gauthier ◽

Celine Charon-Barra ◽

...

Keyword(s):

Breast Cancer ◽

Overall Survival ◽

Neoadjuvant Chemotherapy ◽

Fdg Pet ◽

Metabolic Response ◽

Early Stage ◽

Breast Cancer Subtype ◽

Standardized Uptake Value ◽

Luminal Breast Cancer ◽

Ki 67

11007 Background: To evaluate, in the luminal breast cancer subtype, the prognostic value of tumor glucose metabolism at baseline and of its changes after one cycle of neoadjuvant chemotherapy (NAC). Methods: This prospective study included 61 women with immunophenotypically defined luminal HER2-negative breast cancer treated with NAC. 18F-FDG PET was performed at baseline. Hepatic activity was used as a reference to distinguish between low-metabolic and hypermetabolic tumors. In hypermetabolic tumors, a PET exam was repeated after the first course of NAC. The relative change in the maximal Standardized Uptake Value of the tumor (ΔSUV), corresponding to the metabolic response, was calculated. Results: Forty-two women had hypermetabolic tumors at baseline, corresponding to more proliferative breast cancers with higher Ki-67 expression (p=0.017) and higher grade (p=0.04). Nineteen women had low-metabolic tumors with lower proliferation indexes. Worse overall survival was associated with larger tumor size (>5cm, HR=6.52, P=0.009) and with hypermetabolic tumors achieving a low metabolic response after one cycle of NAC (ΔSUV<16%, HR=10.63, P=0.004). Five-year overall survival in these poor-response patients was 49.22% (95% CI=[14.76%-76.90%]). In contrast, overall survival in women with low-metabolic tumors or hypermetabolic/good-response tumors (ΔSUV≥16%) was good, 100% and 96.15%, respectively (95% CI=[75.69%-99.45%]). Conclusions: In luminal HER2-negative breast tumors, tumor metabolism at baseline and changes after the first course of NAC are surrogate markers of patients’ survival. A subgroup of women with hypermetabolic/bad-responding tumors correlated with poor prognosis can be identified. These results may create the ability to tailor the NAC regimen to the metabolic response at an early stage.

Download Full-text

Genomic Profile of Early Stage Node-Negative Luminal Breast Cancer Associated with Short Disease-Specific Survival

Annals of Oncology ◽

10.1093/annonc/mdu327.31 ◽

2014 ◽

Vol 25 ◽

pp. iv95

Author(s):

A. Durigova ◽

P. Tsantoulis ◽

R. Lyle ◽

C. Borel ◽

G. Fioretta ◽

...

Keyword(s):

Breast Cancer ◽

Early Stage ◽

Luminal Breast Cancer ◽

Genomic Profile ◽

Disease Specific Survival ◽

Node Negative ◽

Disease Specific

Download Full-text

Outcomes after Surgery for Early Stage Breast Cancer in Women Staged With Preoperative Breast Magnetic Resonance Imaging According to Breast Tissue Density

Journal of Breast Imaging ◽

10.1093/jbi/wbz018 ◽

2019 ◽

Vol 1 (2) ◽

pp. 115-121

Author(s):

Renata Faermann ◽

Jonathan Weidenfeld ◽

Leonid Chepelev ◽

Wayne Kendal ◽

Raman Verma ◽

...

Keyword(s):

Breast Cancer ◽

Breast Tissue ◽

Early Stage ◽

Disease Free Survival ◽

Breast Mri ◽

Early Stage Breast Cancer ◽

Free Survival ◽

Tissue Density ◽

Breast Tissue Density ◽

Preoperative Breast Mri

Abstract Purpose To determine surgical outcomes and breast cancer disease-free survival outcomes of women with early stage breast cancer with and without use of preoperative breast MRI according to breast tissue density. Methods Women with early stage breast cancer diagnosed from 2004 to 2009 were classified into 2 groups: 1) those with dense and heterogeneously dense breasts (DB); 2) those with nondense breasts (NDB) (scattered fibroglandular and fatty replaced tissue). The 2 groups were reviewed to determine who underwent preoperative MRI. Breast tissue density was determined with mammography according to ACR BI-RADS. Patients were compared according to tumor size, grade, stage, and treatment. Survival analysis was performed using Kaplan-Meier estimates. Results In total, 261 patients with mean follow-up of 85 months (25–133) were included: 156 DB and 105 NDB. Disease-free survival outcomes were better in the DB group in patients with MRI than in those without MRI: patients with MRI had significantly fewer local recurrences (P < 0.016) and metachronous contralateral breast cancers (P < 0.001), but this was not the case in the NDB group. Mastectomies were higher in the DB group with preoperative MRI than in those without MRI (P < 0.01), as it was in the NDB group (P > 0.05). Conclusions Preoperative breast MRI was associated with reduced local recurrence and metachronous contralateral cancers in the DB group, but not in the NDB group; however, the DB patients with MRI had higher mastectomy rates.

Download Full-text

Abstract P2-05-18: Correlation of breast cancer index (BCI) risk classification with tumor grade and Ki-67 in a large series of patients with early-stage, ER+ breast cancer

10.1158/1538-7445.sabcs16-p2-05-18 ◽

2017 ◽

Author(s):

H Soliman ◽

BE Schroeder ◽

Y Zhang ◽

AM Magliocco ◽

CA Schnabel

Keyword(s):

Breast Cancer ◽

Early Stage ◽

Tumor Grade ◽

Risk Classification ◽

Large Series ◽

Ki 67 ◽

Breast Cancer Index

Download Full-text

Abstract 464: Gene expression signatures for the prediction of endocrine treatment outcome in early-stage luminal breast cancer patients

10.1158/1538-7445.sabcs18-464 ◽

2019 ◽

Author(s):

Werner Schroth ◽

Reiner Hoppe ◽

Florian Büttner ◽

Stefan Winter ◽

Siarhei Kandabarau ◽

...

Keyword(s):

Breast Cancer ◽

Gene Expression ◽

Treatment Outcome ◽

Cancer Patients ◽

Early Stage ◽

Endocrine Treatment ◽

Luminal Breast Cancer ◽

Breast Cancer Patients ◽

Gene Expression Signatures

Download Full-text

Genomic Signatures in Luminal Breast Cancer

Breast Care ◽

10.1159/000509846 ◽

2020 ◽

Vol 15 (4) ◽

pp. 355-365

Author(s):

Julian Puppe ◽

Tabea Seifert ◽

Christian Eichler ◽

Henryk Pilch ◽

Peter Mallmann ◽

...

Keyword(s):

Breast Cancer ◽

High Throughput Sequencing ◽

Early Stage ◽

Intrinsic Subtype ◽

Luminal Breast Cancer ◽

Breast Cancers ◽

Breast Cancer Patients ◽

Cancer Subtypes ◽

Molecular Tests ◽

Immunohistochemical Markers

Background: Breast cancer is a very heterogeneous disease and luminal breast carcinomas represent the hormone receptor-positive tumors among all breast cancer subtypes. In this context, multigene signatures were developed to gain further prognostic and predictive information beyond clinical parameters and traditional immunohistochemical markers. Summary: For early breast cancer patients these molecular tools can guide clinicians to decide on the extension of endocrine therapy to avoid over- and undertreatment by adjuvant chemotherapy. Beside the predictive and prognostic value, a few genomic tests are also able to provide intrinsic subtype classification. In this review, we compare the most frequently used and commercially available molecular tests (OncotypeDX®, MammaPrint®, Prosigna®, EndoPredict®, and Breast Cancer IndexSM). Moreover, we discuss the clinical utility of molecular profiling for advanced breast cancer of the luminal subtype. Key Messages: Multigene assays can help to de-escalate systemic therapy in early-stage breast cancer. Only the Oncotype DX® and MammaPrint®test are validated by entirely prospective and randomized phase 3 trials. More clinical evidence is needed to support the use of genomic tests in node-positive disease. Recent developments in high-throughput sequencing technology will provide further insights to understand the heterogeneity of luminal breast cancers in early-stage and metastatic disease.

Download Full-text

Advanced Imaging and Receipt of Guideline Concordant Care in Women with Early Stage Breast Cancer

International Journal of Breast Cancer ◽

10.1155/2016/2182985 ◽

2016 ◽

Vol 2016 ◽

pp. 1-10 ◽

Cited By ~ 2

Author(s):

Elizabeth Trice Loggers ◽

Diana S. M. Buist ◽

Laura S. Gold ◽

Steven Zeliadt ◽

Rachel Hunter Merrill ◽

...

Keyword(s):

Breast Cancer ◽

Adjuvant Chemotherapy ◽

Early Stage ◽

Breast Mri ◽

Advanced Imaging ◽

Positron Emission ◽

Younger Age ◽

Pet Ct ◽

Estrogen Receptor Positive ◽

Magnetic Resonance Imaging Mri

Objective.It is unknown whether advanced imaging (AI) is associated with higher quality breast cancer (BC) care.Materials and Methods.Claims and Surveillance Epidemiology and End Results data were linked for women diagnosed with incident stage I-III BC between 2002 and 2008 in western Washington State. We examined receipt of preoperative breast magnetic resonance imaging (MRI) or AI (defined as computed tomography [CT]/positron emission tomography [PET]/PET/CT) versus mammogram and/or ultrasound (M-US) alone and receipt of guideline concordant care (GCC) using multivariable logistic regression.Results.Of 5247 women, 67% received M-US, 23% MRI, 8% CT, and 3% PET/PET-CT. In 2002, 5% received MRI and 5% AI compared to 45% and 12%, respectively, in 2008. 79% received GCC, but GCC declined over time and was associated with younger age, urban residence, less comorbidity, shorter time from diagnosis to surgery, and earlier year of diagnosis. Breast MRI was associated with GCC for lumpectomy plus radiation therapy (RT) (OR 1.55, 95% CI 1.08–2.26, andp=0.02) and AI was associated with GCC for adjuvant chemotherapy for estrogen-receptor positive (ER+) BC (OR 1.74, 95% CI 1.17–2.59, andp=0.01).Conclusion.GCC was associated with prior receipt of breast MRI and AI for lumpectomy plus RT and adjuvant chemotherapy for ER+ BC, respectively.

Download Full-text