Preoperative magnetic resonance imaging use and oncologic outcomes in premenopausal breast cancer patients

Northwestern University resource and education development initiatives to advance collaborative artificial intelligence across the learning health system

Introduction

The rapid development of artificial intelligence (AI) in healthcare has exposed the unmet need for growing a multidisciplinary workforce that can collaborate effectively in the learning health systems. Maximizing the synergy among multiple teams is critical for Collaborative AI in Healthcare.

Methods

We have developed a series of data, tools, and educational resources for cultivating the next generation of multidisciplinary workforce for Collaborative AI in Healthcare. We built bulk-natural language processing pipelines to extract structured information from clinical notes and stored them in common data models. We developed multimodal AI/machine learning (ML) tools and tutorials to enrich the toolbox of the multidisciplinary workforce to analyze multimodal healthcare data. We have created a fertile ground to cross-pollinate clinicians and AI scientists and train the next generation of AI health workforce to collaborate effectively.

Results

Our work has democratized access to unstructured health information, AI/ML tools and resources for healthcare, and collaborative education resources. From 2017 to 2022, this has enabled studies in multiple clinical specialties resulting in 68 peer-reviewed publications. In 2022, our cross-discipline efforts converged and institutionalized into the Center for Collaborative AI in Healthcare.

Conclusions

Our Collaborative AI in Healthcare initiatives has created valuable educational and practical resources. They have enabled more clinicians, scientists, and hospital administrators to successfully apply AI methods in their daily research and practice, develop closer collaborations, and advanced the institution-level learning health system.

A Multi-institutional Analysis of Factors Influencing the Rate of Positive MRI Biopsy Among Women with Early-Stage Breast Cancer

BACKGROUND

The use of preoperative magnetic resonance imaging (MRI) for early-stage breast cancer (ESBC) is increasing, but its utility in detecting additional malignancy is unclear and delays surgical management (Jatoi and Benson in Future Oncol 9:347-353, 2013. https://doi.org/10.2217/fon.12.186 , Bleicher et al. J Am Coll Surg 209:180-187, 2009. https://doi.org/10.1016/j.jamcollsurg.2009.04.010 , Borowsky et al. J Surg Res 280:114-122, 2022. https://doi.org/10.1016/j.jss.2022.06.066 ). The present study sought to identify ESBC patients most likely to benefit from preoperative MRI by assessing the positive predictive values (PPVs) of ipsilateral and contralateral biopsies.

METHODS

A retrospective cohort study included patients with cTis-T2N0-N1 breast cancer from two institutions during 2016-2021. A "positive" biopsy result was defined as additional cancer (PositiveCancer) or cancer with histology often excised (PositiveSurg). The PPV of MRI biopsies was calculated with respect to age, family history, breast density, and histology. Uni- and multivariate logistic regression determined whether combinations of age younger than 50 years, dense breasts, family history, and pure ductal carcinoma in situ (DCIS) histology led to higher biopsy yield.

RESULTS

Of the included patients, 447 received preoperative MRI and 131 underwent 149 MRI-guided biopsies (96 ipsilateral, 53 contralateral [18 bilateral]). PositiveCancer for ipsilateral biopsy was 54.2%, and PositiveCancer for contralateral biopsy was 17.0%. PositiveSurg for ipsilateral biopsy was 62.5%, and PositiveSurg for contralateral biopsy was 24.5%. Among the contralateral MRI biopsies, patients younger than 50 years were less likely to have PositiveSurg (odds ratio, 0.02; 95% confidence interval, 0.00-0.84; p = 0.041). The combinations of age, density, family history, and histology did not lead to a higher biopsy yield.

CONCLUSION

Historically accepted factors for recommending preoperative MRI did not appear to confer a higher MRI biopsy yield. To prevent delays to surgical management, MRI should be carefully selected for individual patients most likely to benefit from additional imaging.

miR-125 in Breast Cancer Etiopathogenesis: An Emerging Role as a Biomarker in Differential Diagnosis, Regenerative Medicine, and the Challenges of Personalized Medicine

Breast Cancer (BC) is one of the most common cancer types worldwide, and it is characterized by a complex etiopathogenesis, resulting in an equally complex classification of subtypes. MicroRNA (miRNA or miR) are small non-coding RNA molecules that have an essential role in gene expression and are significantly linked to tumor development and angiogenesis in different types of cancer. Recently, complex interactions among coding and non-coding RNA have been elucidated, further shedding light on the complexity of the roles these molecules fulfill in cancer formation. In this context, knowledge about the role of miR in BC has significantly improved, highlighting the deregulation of these molecules as additional factors influencing BC occurrence, development and classification. A considerable number of papers has been published over the past few years regarding the role of miR-125 in human pathology in general and in several types of cancer formation in particular. Interestingly, miR-125 family members have been recently linked to BC formation as well, and complex interactions (competing endogenous RNA networks, or ceRNET) between this molecule and target mRNA have been described. In this review, we summarize the state-of-the-art about research on this topic.

Breast Magnetic Resonance Imaging for Preoperative Evaluation of Breast Cancer: A Systematic Review and Meta-Analysis

Purpose: Preoperative breast magnetic resonance imaging (MRI) is known to detect additional cancers that are occult on mammography and ultrasound. There is debate as to whether these additional lesions affect clinical outcomes. The objective of this systematic review was to summarize the evidence on whether additional information on disease extent obtained with preoperative breast MRI in patients with newly diagnosed breast cancer affects surgical management, rates of recurrence, survival, re-excision, and early detection of bilateral cancer. Methods: Embase, MEDLINE, and Cochrane Central Register of Controlled Trials were searched until January 2021 (partial update July 2022) for studies comparing outcomes with versus without pre-operative MRI. Included were both randomized controlled trials and other comparative studies provided MRI and control groups had equivalent disease and patient characteristics or methods such as multivariable analysis or propensity score matching were used to control potential confounders. Results: The search resulted in 26,399 citations, of which 8 randomized control trials, 1 prospective cohort study, and 42 retrospective studies met the inclusion criteria. Use of MRI resulted in decreased rates of reoperations (OR = 0.73, 95% CI = 0.63 to 0.85), re-excisions (OR = 0.63, 95% CI = 0.45 to 0.89), and recurrence (HR = 0.77, 95% CI = 0.65 to 0.90). Increased detection of synchronous contralateral breast cancers led to a reduction in metachronous contralateral breast cancer (HR = 0.71, 95% CI = 0.59 to 0.85). Hazard ratios for recurrence-free and overall survival were 0.77 (95% CI = 0.53 to 1.12) and 0.89 (95% CI = 0.74 to 1.07). Conclusion: This systematic review indicates substantial benefits of pre-operative breast MRI in decreasing reoperations and recurrence.

Machine Learning for Lung Cancer Diagnosis, Treatment, and Prognosis

The recent development of imaging and sequencing technologies enables systematic advances in the clinical study of lung cancer. Meanwhile, the human mind is limited in effectively handling and fully utilizing the accumulation of such enormous amounts of data. Machine learning-based approaches play a critical role in integrating and analyzing these large and complex datasets, which have extensively characterized lung cancer through the use of different perspectives from these accrued data. In this review, we provide an overview of machine learning-based approaches that strengthen the varying aspects of lung cancer diagnosis and therapy, including early detection, auxiliary diagnosis, prognosis prediction, and immunotherapy practice. Moreover, we highlight the challenges and opportunities for future applications of machine learning in lung cancer.

The Association of Preoperative Magnetic Resonance Imaging (MRI) With Surgical Management in Patients With Early-Stage Breast Cancer

INTRODUCTION

The rate of mastectomy in lumpectomy-eligible patients with unilateral breast cancer is increasing. We sought to investigate the association between magnetic resonance imaging (MRI) and surgical management of patients with early-stage breast cancer by comparing the rate of mastectomy as first surgery in patients with and without preoperative MRI.

METHODS

A bi-institutional retrospective study included patients diagnosed between 2016 and 2020. Lumpectomy-eligible patients with in situ and invasive cancer were included. Those receiving preoperative therapy, MRI before diagnosis, or with known bilateral cancer were excluded. The risk factors for bilateral and multicentric disease were accounted for. Fisher's exact and chi-square tests compared categorical variables, Wilcoxon two-sample test analyzed continuous variables, and multivariate analyses were performed with Poisson regression.

RESULTS

Four hundred twenty-eight participants met inclusion criteria. Patients who received MRI were younger (58 versus 67 y; P < 0.001) and had denser breasts (group 3 or 4; 61% versus 25%; P < 0.001). Mastectomy rate was twice as high in patients undergoing MRI (32% versus 15%, rate ratio 2.16; P < 0.001), which remained significant in multivariate analysis (rate ratio 2.0; P < 0.001). Contralateral mastectomy (12% versus 4%; P = 0.466) and reexcision (13% versus 12%; P = 0.519) rates were similar. Time to surgery was greater in those receiving MRI alone and MRI biopsy (34 [no MRI] versus 45 [MRI] versus 62 [MRI biopsy]; P < 0.001 for both).

CONCLUSIONS

MRI receipt is associated with a doubled rate of mastectomy in lumpectomy-eligible patients. Future work is needed to standardize patient selection for MRI to those with the highest likelihood of having additional undiagnosed disease.

The clinical impact of MRI on surgical planning for patients with in-breast tumor recurrence

OBJECTIVE

The objective of this study was to evaluate the clinical utility of breast MRI for patients with known in-breast tumor recurrence (IBTR). The aim was to determine if the addition of breast MRI altered surgical approach or multidisciplinary management. Previous studies have focused on using breast MRI for surgical planning for index breast cancers (BC) or detecting IBTR. However, the clinical impact of obtaining MRI in the setting of known IBTR has not been evaluated.

METHODS

A single-institution retrospective chart review was performed to compare surgical approach and multidisciplinary management for patients diagnosed with isolated IBTR who did and did not undergo breast MRI following IBTR diagnosis.

RESULTS

IBTR was identified in 69 patients, 46% of whom underwent MRI. There was no difference in the operative approach (p = 0.14) for IBTR patients who did and did not undergo breast MRI Additionally, there was no difference in multidisciplinary care, treatment order, metastatic disease identification, or mortality between cohorts. A relatively small subgroup of patients (n = 3) required change in surgical plan based on MRI results. Patients proceeding with surgery first who also underwent breast MRI experienced a significantly longer time to surgical intervention (p = 0.03).

CONCLUSION

Breast MRI following IBTR diagnosis infrequently impacted clinical management, including surgical approach and multidisciplinary care. MRI for local disease assessment at the time of IBTR should be used selectively based on clinical concern.

A literature review on the imaging methods for breast cancer

Breast cancer will be easier and more effective to treat if detected early. Breast cancer is assessed and detected using imaging as a primary approach. The capacity to diagnose breast cancers is continually improving thanks to developments in imaging technologies. However, some of these enhancements have been linked to delays in the initiation of treatment procedures of breast cancer. Overall, cancer management relies heavily on imaging procedures such as screening and symptomatic disease management. Mammography, which is considered the gold standard, and breast ultrasonography are employed as routine imaging modalities. Previous research has shown that, despite recent developments, no single imaging modality can detect and characterizing majority of breast lesions. Various imaging methods and their uses in diagnosing and caring the breast cancer are discussed in this study.

Deep learning for cancer type classification and driver gene identification

Background

Genetic information is becoming more readily available and is increasingly being used to predict patient cancer types as well as their subtypes. Most classification methods thus far utilize somatic mutations as independent features for classification and are limited by study power. We aim to develop a novel method to effectively explore the landscape of genetic variants, including germline variants, and small insertions and deletions for cancer type prediction.

Results

We proposed DeepCues, a deep learning model that utilizes convolutional neural networks to unbiasedly derive features from raw cancer DNA sequencing data for disease classification and relevant gene discovery. Using raw whole-exome sequencing as features, germline variants and somatic mutations, including insertions and deletions, were interactively amalgamated for feature generation and cancer prediction. We applied DeepCues to a dataset from TCGA to classify seven different types of major cancers and obtained an overall accuracy of 77.6%. We compared DeepCues to conventional methods and demonstrated a significant overall improvement (p < 0.001). Strikingly, using DeepCues, the top 20 breast cancer relevant genes we have identified, had a 40% overlap with the top 20 known breast cancer driver genes.

Conclusion

Our results support DeepCues as a novel method to improve the representational resolution of DNA sequencings and its power in deriving features from raw sequences for cancer type prediction, as well as discovering new cancer relevant genes.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04400-4.

Magnetic resonance imaging before breast cancer surgery: results of an observational multicenter international prospective analysis (MIPA)

OBJECTIVES

Preoperative breast magnetic resonance imaging (MRI) can inform surgical planning but might cause overtreatment by increasing the mastectomy rate. The Multicenter International Prospective Analysis (MIPA) study investigated this controversial issue.

METHODS

This observational study enrolled women aged 18-80 years with biopsy-proven breast cancer, who underwent MRI in addition to conventional imaging (mammography and/or breast ultrasonography) or conventional imaging alone before surgery as routine practice at 27 centers. Exclusion criteria included planned neoadjuvant therapy, pregnancy, personal history of any cancer, and distant metastases.

RESULTS

Of 5896 analyzed patients, 2763 (46.9%) had conventional imaging only (noMRI group), and 3133 (53.1%) underwent MRI that was performed for diagnosis, screening, or unknown purposes in 692/3133 women (22.1%), with preoperative intent in 2441/3133 women (77.9%, MRI group). Patients in the MRI group were younger, had denser breasts, more cancers ≥ 20 mm, and a higher rate of invasive lobular histology than patients who underwent conventional imaging alone (p < 0.001 for all comparisons). Mastectomy was planned based on conventional imaging in 22.4% (MRI group) versus 14.4% (noMRI group) (p < 0.001). The additional planned mastectomy rate in the MRI group was 11.3%. The overall performed first- plus second-line mastectomy rate was 36.3% (MRI group) versus 18.0% (noMRI group) (p < 0.001). In women receiving conserving surgery, MRI group had a significantly lower reoperation rate (8.5% versus 11.7%, p < 0.001).

CONCLUSIONS

Clinicians requested breast MRI for women with a higher a priori probability of receiving mastectomy. MRI was associated with 11.3% more mastectomies, and with 3.2% fewer reoperations in the breast conservation subgroup.

KEY POINTS

• In 19% of patients of the MIPA study, breast MRI was performed for screening or diagnostic purposes. • The current patient selection to preoperative breast MRI implies an 11% increase in mastectomies, counterbalanced by a 3% reduction of the reoperation rate. • Data from the MIPA study can support discussion in tumor boards when preoperative MRI is under consideration and should be shared with patients to achieve informed decision-making.