Predicting breast cancer response to neoadjuvant treatment using multi-feature MRI: results from the I-SPY 2 TRIAL

Development and validation of an interpretable machine learning model for diagnosing pathologic complete response in breast cancer

BACKGROUND

Pathologic complete response (pCR) following neoadjuvant chemotherapy (NACT) is a critical prognostic marker for patients with breast cancer, potentially allowing surgery omission. However, noninvasive and accurate pCR diagnosis remains a significant challenge due to the limitations of current imaging techniques, particularly in cases where tumors completely disappear post-NACT.

METHODS

We developed a novel framework incorporating Dimensional Accumulation for Layered Images (DALI) and an Attention-Box annotation tool to address the unique challenge of analyzing imaging data where target lesions are absent. These methods transform three-dimensional magnetic resonance imaging into two-dimensional representations and ensure consistent target tracking across time-points. Preprocessing techniques, including tissue-region normalization and subtraction imaging, were used to enhance model performance. Imaging features were extracted using radiomics and pretrained deep-learning models, and machine-learning algorithms were integrated into a stacked ensemble model. The approach was developed using the I-SPY 2 dataset and validated with an independent Tangshan People's Hospital cohort.

RESULTS

The stacked ensemble model achieved superior diagnostic performance, with an area under the receiver operating characteristic curve of 0.831 (95 % confidence interval, 0.769-0.887) on the test set, outperforming individual models. Tissue-region normalization and subtraction imaging significantly enhanced diagnostic accuracy. SHAP analysis identified variables that contributed to the model predictions, ensuring model interpretability.

CONCLUSION

This innovative framework addresses challenges of noninvasive pCR diagnosis. Integrating advanced preprocessing techniques improves feature quality and model performance, supporting clinicians in identifying patients who can safely omit surgery. This innovation reduces unnecessary treatments and improves quality of life for patients with breast cancer.

Role of imaging based response assesment for adapting neoadjuvant systemic therapy for breast cancer: A systematic review

PURPOSE

The objective of this systematic review is to investigate the role of imaging in response monitoring during neoadjuvant systemic therapy (NST) for breast cancer and assess whether treatment modifications based on imaging response are implemented in clinical practice.

METHODS

A systematic review was conducted, analyzing five clinical practice guidelines and 147 clinical trial publications involving NST for breast cancer. The snowballing technique was employed, using a "start set" of clinical guidelines to trace relevant trials. Additionally, a PubMed search was conducted to identify trials published between 2023-2024. The review analyzed the use of imaging modalities, timing, and response criteria, and whether escalation, de-escalation, or change of treatment occurred based on imaging response.

RESULTS

Imaging was utilized in 81 % (119/147) of the trials, with ultrasound, MRI, and mammography being the most frequently employed modalities. Mid-treatment imaging was applied in 56 % (83/147) of the trials. However, only 15 % (22/147) of the trials implemented treatment modifications based on imaging response, highlighting the limited application of imaging response-guided therapy. No standardized imaging protocols or consistent response-guided treatment strategies were identified across the trials or clinical practice guidelines, with considerable variability in imaging methods, timing, and response criteria.

CONCLUSION

This systematic review underscores the critical need for standardized imaging protocols, response assessment criteria and image-guided treatment decisions. It is therefore evident that imaging for response monitoring during treatment should preferably be performed within clinical trials.

Quantifying tumor morphological complexity based on pretreatment MRI fractal analysis for predicting pathologic complete response and survival in breast cancer: a retrospective, multicenter study

BACKGROUND

The tumor morphological complexity is closely associated with treatment response and prognosis in patients with breast cancer. However, conveniently quantifiable tumor morphological complexity methods are currently lacking.

METHODS

Women with breast cancer who underwent NAC and pretreatment MRI were retrospectively enrolled at four centers from May 2010 to April 2023. MRI-based fractal analysis was used to calculate fractal dimensions (FDs), quantifying tumor morphological complexity. Features associated with pCR were identified using multivariable logistic regression analysis, upon which a nomogram model was developed, and assessed by the area under the receiver operating characteristic curve (AUC). Cox proportional hazards analysis was used to identify independent prognostic factors for disease-free survival (DFS) and overall survival (OS) and develop nomogram models.

RESULTS

A total of 1109 patients (median age, 49 years [IQR, 43-54 years]) were included. The training, external validation cohort 1, and cohort 2 included 435, 351, and 323 patients, respectively. HR status (odds ratio [OR], 0.234 [0.135, 0.406]; P < 0.001), HER2 status (OR, 3.320 [1.923, 5.729]; P < 0.001), and Global FD (OR, 0.352 [0.261, 0.480]; P < 0.001) were independent predictors of pCR. The nomogram model for predicting pCR achieved AUCs of 0.80 (95% CI: 0.75, 0.86) and 0.74 (95% CI: 0.68, 0.79) in the external validation cohorts. The nomogram model, which integrated global FD and clinicopathological variables can stratify prognosis into low-risk and high-risk groups (log-rank test, DFS: P = 0.04; OS: P < 0.001).

CONCLUSIONS

Global FD can quantify tumor morphological complexity and the model that combines global FD and clinicopathological variables showed good performance in predicting pCR to NAC and survival in patients with breast cancer.

Nomogram for Predicting Neoadjuvant Chemotherapy Response in Breast Cancer Using MRI-based Intratumoral Heterogeneity Quantification

Background Intratumoral heterogeneity (ITH) in breast cancer contributes to treatment failure and relapse. Noninvasive methods to quantify ITH are currently limited. Purpose To quantify ITH in breast cancer using pretreatment MRI, develop a nomogram to predict pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) and recurrence-free survival (RFS), and investigate biologic pathways associated with nomogram scores. Materials and Methods This retrospective study included patients with breast cancer who underwent NAC at nine centers between April 1988 and December 2023. Tumor regions on MRI scans were clustered and integrated with global pixel distribution patterns to calculate ITH scores. A nomogram for predicting pCR was developed using multivariable logistic regression. A survival dataset was used to evaluate the association between nomogram score and RFS, and a genomics dataset was used to explore the relationship between nomogram score and biologic pathways. Results The study included 1448 women (median age, 49 years [IQR, 43-54 years]). To predict pCR to NAC, the 505 patients from center A served as the training set, and the patients from center B, centers C-F, and center G served as three external validation sets (n = 331, 107, and 384, respectively). The survival set included patients from centers A and H (n = 179), and the genomics set included patients from center I (n = 74). The ITH score was an independent predictor of pCR (odds ratio, 0.12 [95% CI: 0.03, 0.43]; P < .001). The nomogram model achieved area under the receiver operating characteristic curve values of 0.82, 0.81, and 0.79, respectively, in the three external validation sets. A lower nomogram score was correlated with poorer RFS (hazard ratio, 4.04 [95% CI: 1.90, 8.60]; P < .001) and was associated with upregulation of biologic pathways related to tumor proliferation. Conclusion A nomogram model combining ITH score and clinicopathologic variables showed good performance in predicting pCR to NAC and RFS. Published under a CC BY 4.0 license. Supplemental material is available for this article.

Deep Learning Models Based on Pretreatment MRI and Clinicopathological Data to Predict Responses to Neoadjuvant Systemic Therapy in Triple-Negative Breast Cancer

PURPOSE

To develop deep learning models for predicting the pathologic complete response (pCR) to neoadjuvant systemic therapy (NAST) in patients with triple-negative breast cancer (TNBC) based on pretreatment multiparametric breast MRI and clinicopathological data.

METHODS

The prospective institutional review board-approved study [NCT02276443] included 282 patients with stage I-III TNBC who had multiparametric breast MRI at baseline and underwent NAST and surgery during 2016-2021. Dynamic contrast-enhanced MRI (DCE), diffusion-weighted imaging (DWI), and clinicopathological data were used for the model development and internal testing. Data from the I-SPY 2 trial (2010-2016) were used for external testing. Four variables with a potential impact on model performance were systematically investigated: 3D model frameworks, tumor volume preprocessing, tumor ROI selection, and data inputs.

RESULTS

Forty-eight models with different variable combinations were investigated. The best-performing model in the internal testing dataset used DCE, DWI, and clinicopathological data with the originally contoured tumor volume, the tight bounding box of the tumor mask, and ResNeXt50, and achieved an area under the receiver operating characteristic curve (AUC) of 0.76 (95% CI: 0.60-0.88). The best-performing models in the external testing dataset achieved an AUC of 0.72 (95% CI: 0.57-0.84) using only DCE images (originally contoured tumor volume, enlarged bounding box of tumor mask, and ResNeXt50) and an AUC of 0.72 (95% CI: 0.56-0.86) using only DWI images (originally contoured tumor volume, enlarged bounding box of tumor mask, and ResNet18).

CONCLUSIONS

We developed 3D deep learning models based on pretreatment data that could predict pCR to NAST in TNBC patients.

Advances in analytical approaches for background parenchymal enhancement in predicting breast tumor response to neoadjuvant chemotherapy: A systematic review

BACKGROUND

Breast cancer (BC) continues to pose a substantial global health concern, necessitating continuous advancements in therapeutic approaches. Neoadjuvant chemotherapy (NAC) has gained prominence as a key therapeutic strategy, and there is growing interest in the predictive utility of Background Parenchymal Enhancement (BPE) in evaluating the response of breast tumors to NAC. However, the analysis of BPE as a predictive biomarker, along with the techniques used to model BPE changes for accurate and timely predictions of treatment response presents several obstacles. This systematic review aims to thoroughly investigate recent advancements in the analytical methodologies for BPE analysis, and to evaluate their reliability and effectiveness in predicting breast tumor response to NAC, ultimately contributing to the development of personalized and effective therapeutic strategies.

METHODS

A comprehensive and structured literature search was conducted across key electronic databases, including Cochrane Database of Systematic Reviews, Google Scholar, PubMed, and IEEE Xplore covering articles published up to May 10, 2024. The inclusion criteria targeted studies focusing on breast cancer cohorts treated with NAC, involving both pre-treatment and at least one post-treatment breast dynamic contrast-enhanced Magnetic Resonance Imaging (DCE-MRI) scan, and analyzing BPE utility in predicting breast tumor response to NAC. Methodological quality assessment and data extraction were performed to synthesize findings and identify commonalities and differences among various BPE analytical approaches.

RESULTS

The search yielded a total of 882 records. After meticulous screening, 78 eligible records were identified, with 13 studies ultimately meeting the inclusion criteria for the systematic review. Analysis of the literature revealed a significant evolution in BPE analysis, from early studies focusing on single time-point BPE analysis to more recent studies adopting longitudinal BPE analysis. The review uncovered several gaps that compromise the accuracy and timeliness of existing longitudinal BPE analysis methods, such as missing data across multiple imaging time points, manual segmentation of the whole-breast region of interest, and over reliance on traditional statistical methods like logistic regression for modeling BPE and pathological complete response (pCR).

CONCLUSION

This review provides a thorough examination of current advancements in analytical approaches for BPE analysis in predicting breast tumor response to NAC. The shift towards longitudinal BPE analysis has highlighted significant gaps, suggesting the need for alternative analytical techniques, particularly in the realm of artificial intelligence (AI). Future longitudinal BPE research work should focus on standardization in longitudinal BPE measurement and analysis, through integration of deep learning-based approaches for automated tumor segmentation, and implementation of advanced AI technique that can better accommodate varied breast tumor responses, non-linear relationships and complex temporal dynamics in BPE datasets, while also handling missing data more effectively. Such integration could lead to more precise and timely predictions of breast tumor responses to NAC, thereby enhancing personalized and effective breast cancer treatment strategies.

MRI-Based Model for Personalizing Neoadjuvant Treatment in Breast Cancer

BACKGROUND

Functional tumor volume (FTV), measured from dynamic contrast-enhanced MRI, is an imaging biomarker that can predict treatment response in breast cancer patients undergoing neoadjuvant chemotherapy (NAC). The FTV-based predictive model, combined with core biopsy, informed treatment decisions of recommending patients with excellent responses to proceed to surgery early in a large NAC clinical trial.

METHODS

In this retrospective study, we constructed models using FTV measurements. We analyzed performance tradeoffs when a probability threshold was used to identify excellent responders through the prediction of pathology complete response (pCR). Individual models were developed within cohorts defined by the hormone receptor and human epidermal growth factor receptor 2 (HR/HER2) subtype.

RESULTS

A total of 814 patients enrolled in the I-SPY 2 trial between 2010 and 2016 were included with a mean age of 49 years (range: 24 to 77). Among these patients, 289 (36%) achieved pCR. The area under the ROC curve (AUC) ranged from 0.68 to 0.74 for individual HR/HER2 subtypes. When probability thresholds were chosen based on minimum positive predictive value (PPV) levels of 50%, 70%, and 90%, the PPV-sensitivity tradeoff varied among subtypes. The highest sensitivities (100%, 87%, 45%) were found in the HR-/HER2+ sub-cohort for probability thresholds of 0, 0.62, and 0.72; followed by the triple-negative sub-cohort (98%, 52%, 4%) at thresholds of 0.13, 0.58, and 0.67; and HR+/HER2+ (78%, 16%, 8%) at thresholds of 0.34, 0.57, and 0.60. The lowest sensitivities (20%, 0%, 0%) occurred in the HR+/HER2- sub-cohort.

CONCLUSIONS

Predictive models developed using imaging biomarkers, alongside clinically validated probability thresholds, can be incorporated into decision-making for precision oncology.

Comparison of intensity normalization methods in prostate, brain, and breast cancer multi-parametric magnetic resonance imaging

Objectives

Intensity variation in multi-parametric magnetic resonance imaging (MP-MRI) is a confounding factor in MRI analyses. Previous studies have employed several normalization methods, but there is a lack of consensus on which method results in the most comparable images across vendors and acquisitions. This study used MP-MRI collected from patients with confirmed prostate, brain, or breast cancer to examine common intensity normalization methods to identify which best harmonizes intensity values across cofounds.

Materials and methods

Multiple normalization methods were deployed for intensity comparison between three unique sites, MR vendors, and magnetic field strength. Additionally, we calculated radiomic features before and after intensity normalization to determine how downstream analyses may be affected. Specifically, in the prostate cancer cohort, we tested these methods on T2-weighted imaging (T2WI) and additionally looked at a subset of patients who were scanned with and without the use of an endorectal coil (ERC). In a cohort of glioblastoma (GBM) patients, we tested these methods in T1 pre- and post-contrast enhancement (T1, T1C), fluid attenuated inversion recovery (FLAIR), and apparent diffusion coefficient (ADC) maps. Finally, in the breast cancer cohort, we tested methods on T1-weighted nonfat-suppressed images. All methods were compared using a two one-sided test (TOST) to test for equivalence of mean and standard deviation of intensity distributions.

Results

While each organ had unique results, across every tested comparison, using the Z-score of intensity within a mask of the organ consistently provided an equivalent distribution (all p < 0.001).

Conclusions

Our results suggest that intensity normalization using the Z-score of intensity within prostate, breast, and brain MR images produces the most comparable intensities between sites, MR vendors, magnetic field strength, and prostate endorectal coil usage. Likewise, Z-score normalization provided the highest percentage of radiomic features that were statistically equal across the three organs.

Patient-Derived Xenografts of Triple-Negative Breast Cancer Enable Deconvolution and Prediction of Chemotherapy Responses

Combination chemotherapy remains essential for clinical management of triple-negative breast cancer (TNBC). Consequently, responses to multiple single agents cannot be delineated at the single patient level, even though some patients might not require all drugs in the combination. Herein, we conduct multi-omic analyses of orthotopic TNBC patient-derived xenografts (PDXs) treated with single agent carboplatin, docetaxel, or the combination. Combination responses were usually no better than the best single agent, with enhanced response in only ~13% of PDX, and apparent antagonism in a comparable percentage. Single-omic comparisons showed largely non-overlapping results between genes associated with single agent and combination treatments that could be validated in independent patient cohorts. Multi-omic analyses of PDXs identified agent-specific biomarkers/biomarker combinations, nominating high Cytokeratin-5 (KRT5) as a general marker of responsiveness. Notably, integrating proteomic with transcriptomic data improved predictive modeling of pathologic complete response to combination chemotherapy. PDXs refractory to all treatments were enriched for signatures of dysregulated mitochondrial function. Targeting this process indirectly in a PDX with HDAC inhibition plus chemotherapy in vivo overcomes chemoresistance. These results suggest possible resistance mechanisms and therapeutic strategies in TNBC to overcome chemoresistance, and potentially allow optimization of chemotherapeutic regimens.

Amplitude-Modulation Frequency Optimization for Enhancing Harmonic Motion Imaging Performance of Breast Tumors in the Clinic

OBJECTIVE

Elastography images tissue mechanical responses and infers the underlying properties to aid diagnosis and treatment response monitoring. The estimation of absolute or relative tumor properties may vary with dimensions even when the mechanical properties remain constant. Harmonic motion imaging (HMI) uses amplitude-modulated (AM) focused ultrasound to interrogate the targeted tissue's viscoelastic properties. In this study, effects of AM frequencies on HMI were investigated in terms of inclusion relative stiffness and size estimation.

METHODS

AM frequencies from 200 to 600 Hz in steps of 100 Hz were considered using a 5.3-kPa phantom with cylindrical inclusions (Young's modulus: 22, 31, 44, 56 kPa, and diameter: 4.8, 8.1, 13.6, 19.8 mm) to optimize the performance of HMI in characterizing tumors with the same mechanical properties and of different dimensions.

RESULTS

Consistent displacement ratios (DRs) (17.5% variation) of the inclusion to background were obtained with 200-Hz AM for breast-tumor-mimicking inclusions albeit a suboptimal inclusion size estimation obtained. 400-Hz was otherwise used for small and low-contrast inclusions (4.8 mm, 22 or 31 kPa). A linear relationship (R2 = 0.9043) was found between the inverse DR at these frequencies and the Young's modulus ratio. 400 Hz obtained the most accurate inclusion size estimation with an overall estimation error on the lateral dimension of 0.5 mm. In vivo imaging of breast cancer patients (n = 5) was performed at 200 or 400 Hz.

CONCLUSION

The results presented herein indicate that the HMI AM frequency could be optimized adaptively in cases of different applications, i.e., at 200 or 400 Hz, depending on whether aimed for consistent DR measurement for tumor response assessment or tumor margin delineation for surgical planning. HMI may thus be capable of predicting the pathologic endpoint of tumors in response to neoadjuvant chemotherapy (NACT) as early as 3 weeks into treatment.

Time-Series MR Images Identifying Complete Response to Neoadjuvant Chemotherapy in Breast Cancer Using a Deep Learning Approach

BACKGROUND

Pathological complete response (pCR) is an essential criterion for adjusting follow-up treatment plans for patients with breast cancer (BC). The value of the visual geometry group and long short-term memory (VGG-LSTM) network using time-series dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for pCR identification in BC is unclear.

PURPOSE

To identify pCR to neoadjuvant chemotherapy (NAC) using deep learning (DL) models based on the VGG-LSTM network.

STUDY TYPE

Retrospective.

POPULATION

Center A: 235 patients (47.7 ± 10.0 years) were divided 7:3 into training (n = 164) and validation set (n = 71). Center B: 150 patients (48.5 ± 10.4 years) were used as test set.

FIELD STRENGTH/SEQUENCE

3-T, T2-weighted spin-echo sequence imaging, and gradient echo DCE sequence imaging.

ASSESSMENT

Patients underwent MRI examinations at three sequential time points: pretreatment, after three cycles of treatment, and prior to surgery, with tumor regions of interest manually delineated. Histopathology was the gold standard. We used VGG-LSTM network to establish seven DL models using time-series DCE-MR images: pre-NAC images (t0 model), early NAC images (t1 model), post-NAC images (t2 model), pre-NAC and early NAC images (t0 + t1 model), pre-NAC and post-NAC images (t0 + t2 model), pre-NAC, early NAC and post-NAC images (t0 + t1 + t2 model), and the optimal model combined with the clinical features and imaging features (combined model). The models were trained and optimized on the training and validation set, and tested on the test set.

STATISTICAL TESTS

The DeLong, Student's t-test, Mann-Whitney U, Chi-squared, Fisher's exact, Hosmer-Lemeshow tests, decision curve analysis, and receiver operating characteristics analysis were performed. P < 0.05 was considered significant.

RESULTS

Compared with the other six models, the combined model achieved the best performance in the test set yielding an AUC of 0.927.

DATA CONCLUSION

The combined model that used time-series DCE-MR images, clinical features and imaging features shows promise for identifying pCR in BC.

LEVEL OF EVIDENCE: 4

TECHNICAL EFFICACY

Stage 4.

Datopotamab-deruxtecan plus durvalumab in early-stage breast cancer: the sequential multiple assignment randomized I-SPY2.2 phase 2 trial

Sequential adaptive trial designs can help accomplish the goals of personalized medicine, optimizing outcomes and avoiding unnecessary toxicity. Here we describe the results of incorporating a promising antibody-drug conjugate, datopotamab-deruxtecan (Dato-DXd) in combination with programmed cell death-ligand 1 inhibitor, durvalumab, as the first sequence of therapy in the I-SPY2.2 phase 2 neoadjuvant sequential multiple assignment randomization trial for high-risk stage 2/3 breast cancer. The trial includes three blocks of treatment, with initial randomization to different experimental agent(s) (block A), followed by a taxane-based regimen tailored to tumor subtype (block B), followed by doxorubicin-cyclophosphamide (block C). Subtype-specific algorithms based on magnetic resonance imaging volume change and core biopsy guide treatment redirection after each block, including the option of early surgical resection in patients predicted to have a high likelihood of pathologic complete response, which is the primary endpoint assessed when resection occurs. There are two primary efficacy analyses: after block A and across all blocks for six prespecified HER2-negative subtypes (defined by hormone receptor status and/or response-predictive subtypes). In total, 106 patients were treated with Dato-DXd/durvalumab in block A. In the immune-positive subtype, Dato-DXd/durvalumab exceeded the prespecified threshold for success (graduated) after block A; and across all blocks, pathologic complete response rates were equivalent to the rate expected for the standard of care (79%), but 54% achieved that result after Dato-DXd/durvalumab alone (block A) and 92% without doxorubicin-cyclophosphamide (after blocks A + B). The treatment strategy across all blocks graduated in the hormone-negative/immune-negative subtype. No new toxicities were observed. Stomatitis was the most common side effect in block A. No patients receiving block A treatment alone had adrenal insufficiency. Dato-DXd/durvalumab is a promising therapy combination that can eliminate standard chemotherapy in many patients, particularly the immune-positive subtype.ClinicalTrials.gov registration: NCT01042379 .

Diffusion-weighted imaging in addition to contrast-enhanced MRI in identifying complete response in HER2-positive breast cancer

OBJECTIVES

The aim of this study is to investigate the added value of diffusion-weighted imaging (DWI) to dynamic-contrast enhanced (DCE)-MRI to identify a pathological complete response (pCR) in patients with HER2-positive breast cancer and radiological complete response (rCR).

MATERIALS AND METHODS

This is a single-center observational study of 102 patients with stage I-III HER2-positive breast cancer and real-world documented rCR on DCE-MRI. Patients were treated between 2015 and 2019. Both 1.5 T/3.0 T single-shot diffusion-weighted echo-planar sequence were used. Post neoadjuvant systemic treatment (NST) diffusion-weighted images were reviewed by two readers for visual evaluation and ADCmean. Discordant cases were resolved in a consensus meeting. pCR of the breast (ypT0/is) was used to calculate the negative predictive value (NPV). Breast pCR-percentages were tested with Fisher's exact test. ADCmean and ∆ADCmean(%) for patients with and without pCR were compared using a Mann-Whitney U-test.

RESULTS

The NPV for DWI added to DCE is 86% compared to 87% for DCE alone in hormone receptor (HR)-/HER2-positive and 67% compared to 64% in HR-positive/HER2-positive breast cancer. Twenty-seven of 39 non-rCR DWI cases were false positives. In HR-positive/HER2-positive breast cancer the NPV for DCE MRI differs between MRI field strength (1.5 T: 50% vs. 3 T: 81% [p = 0.02]). ADCmean at baseline, post-NST, and ∆ADCmean were similar between patients with and without pCR.

CONCLUSION

DWI has no clinically relevant effect on the NPV of DCE alone to identify a pCR in early HER2-positive breast cancer. The added value of DWI in HR-positive/HER2-positive breast cancer should be further investigated taken MRI field strength into account.

CLINICAL RELEVANCE STATEMENT

The residual signal on DWI after neoadjuvant systemic therapy in cases with early HER2-positive breast cancer and no residual pathologic enhancement on DCE-MRI breast should not (yet) be considered in assessing a complete radiologic response.

KEY POINTS

Radiologic complete response is associated with a pathologic complete response (pCR) in HER2+ breast cancer but further improvement is warranted. No relevant increase in negative predictive value was observed when DWI was added to DCE. Residual signal on DW-images without pathologic enhancement on DCE-MRI, does not indicate a lower chance of pCR.

Datopotamab-deruxtecan in early-stage breast cancer: the sequential multiple assignment randomized I-SPY2.2 phase 2 trial

Among the goals of patient-centric care are the advancement of effective personalized treatment, while minimizing toxicity. The phase 2 I-SPY2.2 trial uses a neoadjuvant sequential therapy approach in breast cancer to further these goals, testing promising new agents while optimizing individual outcomes. Here we tested datopotamab-deruxtecan (Dato-DXd) in the I-SPY2.2 trial for patients with high-risk stage 2/3 breast cancer. I-SPY2.2 uses a sequential multiple assignment randomization trial design that includes three sequential blocks of biologically targeted neoadjuvant treatment: the experimental agent(s) (block A), a taxane-based regimen tailored to the tumor subtype (block B) and doxorubicin-cyclophosphamide (block C). Patients are randomized into arms consisting of different investigational block A treatments. Algorithms based on magnetic resonance imaging and core biopsy guide treatment redirection after each block, including the option of early surgical resection in patients predicted to have a high likelihood of pathological complete response, the primary endpoint. There are two primary efficacy analyses: after block A and across all blocks for the six prespecified breast cancer subtypes (defined by clinical hormone receptor/human epidermal growth factor receptor 2 (HER2) status and/or the response-predictive subtypes). We report results of 103 patients treated with Dato-DXd. While Dato-DXd did not meet the prespecified threshold for success (graduation) after block A in any subtype, the treatment strategy across all blocks graduated in the hormone receptor-negative HER2-Immune-DNA repair deficiency- subtype with an estimated pathological complete response rate of 41%. No new toxicities were observed, with stomatitis and ocular events occurring at low grades. Dato-DXd was particularly active in the hormone receptor-negative/HER2-Immune-DNA repair deficiency- signature, warranting further investigation, and was safe in other subtypes in patients who followed the treatment strategy. ClinicalTrials.gov registration: NCT01042379 .

Tumor Morphology for Prediction of Poor Responses Early in Neoadjuvant Chemotherapy for Breast Cancer: A Multicenter Retrospective Study

BACKGROUND

This multicenter and retrospective study investigated the additive value of tumor morphologic features derived from the functional tumor volume (FTV) tumor mask at pre-treatment (T0) and the early treatment time point (T1) in the prediction of pathologic outcomes for breast cancer patients undergoing neoadjuvant chemotherapy.

METHODS

A total of 910 patients enrolled in the multicenter I-SPY 2 trial were included. FTV and tumor morphologic features were calculated from the dynamic contrast-enhanced (DCE) MRI. A poor response was defined as a residual cancer burden (RCB) class III (RCB-III) at surgical excision. The area under the receiver operating characteristic curve (AUC) was used to evaluate the predictive performance. The analysis was performed in the full cohort and in individual sub-cohorts stratified by hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) status.

RESULTS

In the full cohort, the AUCs for the use of the FTV ratio and clinicopathologic data were 0.64 ± 0.03 (mean ± SD [standard deviation]). With morphologic features, the AUC increased significantly to 0.76 ± 0.04 (p < 0.001). The ratio of the surface area to volume ratio between T0 and T1 was found to be the most contributing feature. All top contributing features were from T1. An improvement was also observed in the HR+/HER2- and triple-negative sub-cohorts. The AUC increased significantly from 0.56 ± 0.05 to 0.70 ± 0.06 (p < 0.001) and from 0.65 ± 0.06 to 0.73 ± 0.06 (p < 0.001), respectively, when adding morphologic features.

CONCLUSION

Tumor morphologic features can improve the prediction of RCB-III compared to using FTV only at the early treatment time point.

Diffusion Tensor Imaging for Characterizing Changes in Triple-Negative Breast Cancer During Neoadjuvant Systemic Therapy

BACKGROUND

Assessment of treatment response in triple-negative breast cancer (TNBC) may guide individualized care for improved patient outcomes. Diffusion tensor imaging (DTI) measures tissue anisotropy and could be useful for characterizing changes in the tumors and adjacent fibroglandular tissue (FGT) of TNBC patients undergoing neoadjuvant systemic treatment (NAST).

PURPOSE

To evaluate the potential of DTI parameters for prediction of treatment response in TNBC patients undergoing NAST.

STUDY TYPE

Prospective.

POPULATION

Eighty-six women (average age: 51 ± 11 years) with biopsy-proven clinical stage I-III TNBC who underwent NAST followed by definitive surgery. 47% of patients (40/86) had pathologic complete response (pCR).

FIELD STRENGTH/SEQUENCE

3.0 T/reduced field of view single-shot echo-planar DTI sequence.

ASSESSMENT

Three MRI scans were acquired longitudinally (pre-treatment, after 2 cycles of NAST, and after 4 cycles of NAST). Eleven histogram features were extracted from DTI parameter maps of tumors, a peritumoral region (PTR), and FGT in the ipsilateral breast. DTI parameters included apparent diffusion coefficients and relative diffusion anisotropies. pCR status was determined at surgery.

STATISTICAL TESTS

Longitudinal changes of DTI features were tested for discrimination of pCR using Mann-Whitney U test and area under the receiver operating characteristic curve (AUC). A P value <0.05 was considered statistically significant.

RESULTS

47% of patients (40/86) had pCR. DTI parameters assessed after 2 and 4 cycles of NAST were significantly different between pCR and non-pCR patients when compared between tumors, PTRs, and FGTs. The median surface/average anisotropy of the PTR, measured after 2 and 4 cycles of NAST, increased in pCR patients and decreased in non-pCR patients (AUC: 0.78; 0.027 ± 0.043 vs. -0.017 ± 0.042 mm2/s).

DATA CONCLUSION

Quantitative DTI features from breast tumors and the peritumoral tissue may be useful for predicting the response to NAST in TNBC.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 4.

Identifying radiogenomic associations of breast cancer based on DCE-MRI by using Siamese Neural Network with manufacturer bias normalization

BACKGROUND AND PURPOSE

The immunohistochemical test (IHC) for Human Epidermal Growth Factor Receptor 2 (HER2) and hormone receptors (HR) provides prognostic information and guides treatment for patients with invasive breast cancer. The objective of this paper is to establish a non-invasive system for identifying HER2 and HR in breast cancer using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI).

METHODS

In light of the absence of high-performance algorithms and external validation in previously published methods, this study utilizes 3D deep features and radiomics features to represent the information of the Region of Interest (ROI). A Siamese Neural Network was employed as the classifier, with 3D deep features and radiomics features serving as the network input. To neutralize manufacturer bias, a batch effect normalization method, ComBat, was introduced. To enhance the reliability of the study, two datasets, Predict Your Therapeutic Response with Imaging and moLecular Analysis (I-SPY 1) and I-SPY 2, were incorporated. I-SPY 2 was utilized for model training and validation, while I-SPY 1 was exclusively employed for external validation. Additionally, a breast tumor segmentation network was trained to improve radiomic feature extraction.

RESULTS

The results indicate that our approach achieved an average Area Under the Curve (AUC) of 0.632, with a Standard Error of the Mean (SEM) of 0.042 for HER2 prediction in the I-SPY 2 dataset. For HR prediction, our method attained an AUC of 0.635 (SEM 0.041), surpassing other published methods in the AUC metric. Moreover, the proposed method yielded competitive results in other metrics. In external validation using the I-SPY 1 dataset, our approach achieved an AUC of 0.567 (SEM 0.032) for HR prediction and 0.563 (SEM 0.033) for HER2 prediction.

CONCLUSION

This study proposes a non-invasive system for identifying HER2 and HR in breast cancer. Although the results do not conclusively demonstrate superiority in both tasks, they indicate that the proposed method achieved good performance and is a competitive classifier compared to other reference methods. Ablation studies demonstrate that both radiomics features and deep features for the Siamese Neural Network are beneficial for the model. The introduced manufacturer bias normalization method has been shown to enhance the method's performance. Furthermore, the external validation of the method enhances the reliability of this research. Source code, pre-trained segmentation network, Radiomics and deep features, data for statistical analysis, and Supporting Information of this article are online at: https://github.com/FORRESTHUACHEN/Siamese_Neural_Network_based_Brest_cancer_Radiogenomic.

Use of Pretreatment Perfusion MRI-based Intratumoral Heterogeneity to Predict Pathologic Response of Triple-Negative Breast Cancer to Neoadjuvant Chemoimmunotherapy

Background Neoadjuvant chemoimmunotherapy (NACI) has significantly increased the rate of pathologic complete response (pCR) in patients with early-stage triple-negative breast cancer (TNBC), although predictors of response to this regimen have not been identified. Purpose To investigate pretreatment perfusion MRI-based radiomics as a predictive marker for pCR in patients with TNBC undergoing NACI. Materials and Methods This prospective study enrolled women with early-stage TNBC who underwent NACI at two different centers from August 2021 to July 2023. Pretreatment dynamic contrast-enhanced MRI scans obtained using scanners from multiple vendors were analyzed using the Tofts model to segment tumors and analyze pharmacokinetic parameters. Radiomics features were extracted from the rate constant for contrast agent plasma-to-interstitial transfer (or Ktrans), volume fraction of extravascular and extracellular space (Ve), and maximum contrast agent uptake rate (Slopemax) maps and analyzed using unsupervised correlation and least absolute shrinkage and selector operator, or LASSO, to develop a radiomics score. Score effectiveness was assessed using the area under the receiver operating characteristic curve (AUC), and multivariable logistic regression was used to develop a multimodal nomogram for enhanced prediction. The discrimination, calibration, and clinical utility of the nomogram were evaluated in an external test set. Results The training set included 112 female participants from center 1 (mean age, 52 years ± 11 [SD]), and the external test set included 83 female participants from center 2 (mean age, 47 years ± 11). The radiomics score demonstrated an AUC of 0.80 (95% CI: 0.70, 0.89) for predicting pCR. A nomogram incorporating the radiomics score, grade, and Ki-67 yielded an AUC of 0.86 (95% CI: 0.78, 0.94) in the test set. Associations were found between higher radiomics score (>0.25) and tumor size (P < .001), washout enhancement (P = .01), androgen receptor expression (P = .009), and programmed death ligand 1 expression (P = .01), demonstrating a correlation with tumor immune environment in participants with TNBC. Conclusion A radiomics score derived from pharmacokinetic parameters at pretreatment dynamic contrast-enhanced MRI exhibited good performance for predicting pCR in participants with TNBC undergoing NACI, and could potentially be used to enhance clinical decision making. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Rauch in this issue.

Comprehensive characterization of tumor therapeutic response with simultaneous mapping cell size, density, and transcytolemmal water exchange

Early assessment of tumor therapeutic response is an important topic in precision medicine to optimize personalized treatment regimens and reduce unnecessary toxicity, cost, and delay. Although diffusion MRI (dMRI) has shown potential to address this need, its predictive accuracy is limited, likely due to its unspecific sensitivity to overall pathological changes. In this work, we propose a new quantitative dMRI-based method dubbed EXCHANGE (MRI of water Exchange, Confined and Hindered diffusion under Arbitrary Gradient waveform Encodings) for simultaneous mapping of cell size, cell density, and transcytolemmal water exchange. Such rich microstructural information comprehensively evaluates tumor pathologies at the cellular level. Validations using numerical simulations and in vitro cell experiments confirmed that the EXCHANGE method can accurately estimate mean cell size, density, and water exchange rate constants. The results from in vivo animal experiments show the potential of EXCHANGE for monitoring tumor treatment response. Finally, the EXCHANGE method was implemented in breast cancer patients with neoadjuvant chemotherapy, demonstrating its feasibility in assessing tumor therapeutic response in clinics. In summary, a new, quantitative dMRI-based EXCHANGE method was proposed to comprehensively characterize tumor microstructural properties at the cellular level, suggesting a unique means to monitor tumor treatment response in clinical practice.

3-year invasive disease-free survival with chemotherapy de-escalation using an 18F-FDG-PET-based, pathological complete response-adapted strategy in HER2-positive early breast cancer (PHERGain): a randomised, open-label, phase 2 trial

BACKGROUND

PHERGain was designed to assess the feasibility, safety, and efficacy of a chemotherapy-free treatment based on a dual human epidermal growth factor receptor 2 (HER2) blockade with trastuzumab and pertuzumab in patients with HER2-positive early breast cancer (EBC). It used an 18fluorine-fluorodeoxyglucose-PET-based, pathological complete response (pCR)-adapted strategy.

METHODS

PHERGain was a randomised, open-label, phase 2 trial that took place in 45 hospitals in seven European countries. It randomly allocated patients in a 1:4 ratio with centrally confirmed, HER2-positive, stage I-IIIA invasive, operable breast cancer with at least one PET-evaluable lesion to either group A, where patients received docetaxel (75 mg/m2, intravenous), carboplatin (area under the curve 6 mg/mL per min, intravenous), trastuzumab (600 mg fixed dose, subcutaneous), and pertuzumab (840 mg loading dose followed by 420 mg maintenance doses, intravenous; TCHP), or group B, where patients received trastuzumab and pertuzumab with or without endocrine therapy, every 3 weeks. Random allocation was stratified by hormone receptor status. Centrally reviewed PET was conducted at baseline and after two treatment cycles. Patients in group B were treated according to on-treatment PET results. Patients in group B who were PET-responders continued with trastuzumab and pertuzumab with or without endocrine therapy for six cycles, while PET-non-responders were switched to receive six cycles of TCHP. After surgery, patients in group B who were PET-responders who did not achieve a pCR received six cycles of TCHP, and all patients completed up to 18 cycles of trastuzumab and pertuzumab. The primary endpoints were pCR in patients who were group B PET-responders after two treatment cycles (the results for which have been reported previously) and 3-year invasive disease-free survival (iDFS) in patients in group B. The study is registered with ClinicalTrials.gov (NCT03161353) and is ongoing.

FINDINGS

Between June 26, 2017, and April 24, 2019, a total of 356 patients were randomly allocated (71 patients in group A and 285 patients in group B), and 63 (89%) and 267 (94%) patients proceeded to surgery in groups A and B, respectively. At this second analysis (data cutoff: Nov 4, 2022), the median duration of follow-up was 43·3 months (range 0·0-63·0). In group B, the 3-year iDFS rate was 94·8% (95% CI 91·4-97·1; p=0·001), meeting the primary endpoint. No new safety signals were identified. Treatment-related adverse events and serious adverse events (SAEs) were numerically higher in patients allocated to group A than to group B (grade ≥3 62% vs 33%; SAEs 28% vs 14%). Group B PET-responders with pCR presented the lowest incidence of treatment-related grade 3 or higher adverse events (1%) without any SAEs.

INTERPRETATION

Among HER2-positive EBC patients, a PET-based, pCR-adapted strategy was associated with an excellent 3-year iDFS. This strategy identified about a third of patients who had HER2-positive EBC who could safely omit chemotherapy.

FUNDING

F Hoffmann-La Roche.

SwinHR: Hemodynamic-powered hierarchical vision transformer for breast tumor segmentation

Accurate and automated segmentation of breast tumors in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) plays a critical role in computer-aided diagnosis and treatment of breast cancer. However, this task is challenging, due to random variation in tumor sizes, shapes, appearances, and blurred boundaries of tumors caused by inherent heterogeneity of breast cancer. Moreover, the presence of ill-posed artifacts in DCE-MRI further complicate the process of tumor region annotation. To address the challenges above, we propose a scheme (named SwinHR) integrating prior DCE-MRI knowledge and temporal-spatial information of breast tumors. The prior DCE-MRI knowledge refers to hemodynamic information extracted from multiple DCE-MRI phases, which can provide pharmacokinetics information to describe metabolic changes of the tumor cells over the scanning time. The Swin Transformer with hierarchical re-parameterization large kernel architecture (H-RLK) can capture long-range dependencies within DCE-MRI while maintaining computational efficiency by a shifted window-based self-attention mechanism. The use of H-RLK can extract high-level features with a wider receptive field, which can make the model capture contextual information at different levels of abstraction. Extensive experiments are conducted in large-scale datasets to validate the effectiveness of our proposed SwinHR scheme, demonstrating its superiority over recent state-of-the-art segmentation methods. Also, a subgroup analysis split by MRI scanners, field strength, and tumor size is conducted to verify its generalization. The source code is released on (https://github.com/GDPHMediaLab/SwinHR).

Deep Learning-Based Segmentation of Locally Advanced Breast Cancer on MRI in Relation to Residual Cancer Burden: A Multi-Institutional Cohort Study

BACKGROUND

While several methods have been proposed for automated assessment of breast-cancer response to neoadjuvant chemotherapy on breast MRI, limited information is available about their performance across multiple institutions.

PURPOSE

To assess the value and robustness of deep learning-derived volumes of locally advanced breast cancer (LABC) on MRI to infer the presence of residual disease after neoadjuvant chemotherapy.

STUDY TYPE

Retrospective.

SUBJECTS

Training cohort: 102 consecutive female patients with LABC scheduled for neoadjuvant chemotherapy (NAC) from a single institution (age: 25-73 years). Independent testing cohort: 55 consecutive female patients with LABC from four institutions (age: 25-72 years).

FIELD STRENGTH/SEQUENCE

Training cohort: single vendor 1.5 T or 3.0 T. Testing cohort: multivendor 3.0 T. Gradient echo dynamic contrast-enhanced sequences.

ASSESSMENT

A convolutional neural network (nnU-Net) was trained to segment LABC. Based on resulting tumor volumes, an extremely randomized tree model was trained to assess residual cancer burden (RCB)-0/I vs. RCB-II/III. An independent model was developed using functional tumor volume (FTV). Models were tested on an independent testing cohort and response assessment performance and robustness across multiple institutions were assessed.

STATISTICAL TESTS

The receiver operating characteristic (ROC) was used to calculate the area under the ROC curve (AUC). DeLong's method was used to compare AUCs. Correlations were calculated using Pearson's method. P values <0.05 were considered significant.

RESULTS

Automated segmentation resulted in a median (interquartile range [IQR]) Dice score of 0.87 (0.62-0.93), with similar volumetric measurements (R = 0.95, P < 0.05). Automated volumetric measurements were significantly correlated with FTV (R = 0.80). Tumor volume-derived from deep learning of DCE-MRI was associated with RCB, yielding an AUC of 0.76 to discriminate between RCB-0/I and RCB-II/III, performing similar to the FTV-based model (AUC = 0.77, P = 0.66). Performance was comparable across institutions (IQR AUC: 0.71-0.84).

DATA CONCLUSION

Deep learning-based segmentation estimates changes in tumor load on DCE-MRI that are associated with RCB after NAC and is robust against variations between institutions.

EVIDENCE LEVEL

2.

TECHNICAL EFFICACY

Stage 4.

Bilateral asymmetry of quantitative parenchymal kinetics at ultrafast DCE-MRI predict response to neoadjuvant chemotherapy in patients with HER2+ breast cancer

PURPOSE

To assess whether measurement of the bilateral asymmetry of semiquantitative and quantitative perfusion parameters from ultrafast dynamic contrast-enhanced MRI (DCE-MRI), allows early prediction of pathologic response after neoadjuvant chemotherapy (NAC) in patients with HER2+ breast cancer.

MATERIALS AND METHODS

Twenty-eight female patients with HER2+ breast cancer treated with NAC who underwent pre-NAC ultrafast DCE-MRI (3-9 s/phase) were enrolled for this study. Four semiquantitative and two quantitative parenchymal parameters were calculated for each patient. Ipsilateral/contralateral (I/C) ratio (for four parameters) and the difference between (for two parameters) ipsi- and contra-lateral parenchymal kinetic parameters (kBPE) were compared for patients with pathologic complete response (pCR) and those having residual disease. Lasso regression with leave-one-out cross validation was used to determine the optimal combination of parameters for a regression model and multivariable logistic regression was used to identify independent predictors for pCR. Chi-squared test, two-sided t-test and Kruskal-Wallis test were used.

RESULTS

The Ktrans I/C ratio cutoff value of 1.11 had a sensitivity of 83.3% and specificity of 75% for pCR. The ve I/C ratio cutoff value of 1.1 had a sensitivity of 75% and specificity of 81.3% for pCR. The area under the receiver operating characteristic curve of the three-kBPE parameter model, including initial area under the enhancement curve (AUC30) I/C ratio, KtransI/C ratio and ve I/C ratio, was 0.89 with sensitivity of 91.7% at specificity of 81.3%.

CONCLUSION

Quantitative assessment of bilateral asymmetry kBPE from pre-NAC ultrafast DCE-MRI can predict pCR in patients with HER2+ breast cancer.

Exploring Neoadjuvant Chemotherapy, Predictive Models, Radiomic, and Pathological Markers in Breast Cancer: A Comprehensive Review

Breast cancer retains its position as the most prevalent form of malignancy among females on a global scale. The careful selection of appropriate treatment for each patient holds paramount importance in effectively managing breast cancer. Neoadjuvant chemotherapy (NACT) plays a pivotal role in the comprehensive treatment of this disease. Administering chemotherapy before surgery, NACT becomes a powerful tool in reducing tumor size, potentially enabling fewer invasive surgical procedures and even rendering initially inoperable tumors amenable to surgery. However, a significant challenge lies in the varying responses exhibited by different patients towards NACT. To address this challenge, researchers have focused on developing prediction models that can identify those who would benefit from NACT and those who would not. Such models have the potential to reduce treatment costs and contribute to a more efficient and accurate management of breast cancer. Therefore, this review has two objectives: first, to identify the most effective radiomic markers correlated with NACT response, and second, to explore whether integrating radiomic markers extracted from radiological images with pathological markers can enhance the predictive accuracy of NACT response. This review will delve into addressing these research questions and also shed light on the emerging research direction of leveraging artificial intelligence techniques for predicting NACT response, thereby shaping the future landscape of breast cancer treatment.

Changing the role of pCR in breast cancer treatment - an unjustifiable interpretation of a good prognostic factor as a "factor for a good prognosis"

Pathologic complete response (pCR) after neoadjuvant systemic therapy (NAST) of early breast cancer (EBC) has been recognized as a good prognostic factor in the treatment of breast cancer because of its significant correlation with long-term disease outcome. Based on this correlation, pCR has been accepted by health authorities (FDA, EMA) as a surrogate endpoint in clinical trials for accelerated drug approval. Moreover, in recent years, we have observed a tendency to treat pCR in routine clinical practice as a primary therapeutic target rather than just one of the pieces of information obtained from clinical trials. These trends in routine clinical practice are the result of recommendations in treatment guidelines, such as the ESMO recommendation "…to deliver all planned (neoadjuvant) treatment without unnecessary breaks, i.e. without dividing it into preoperative and postoperative periods, irrespective of the magnitude of tumor response", because "…this will increase the probability of achieving pCR, which is a proven factor for a good prognosis…". We hypothesize that the above recommendations and trends in routine clinical practice are the consequences of misunderstanding regarding the concept of pCR, which has led to a shift in its importance from a prognostic factor to a desired treatment outcome. The origin of this misunderstanding could be a strong subconscious incentive to achieve pCR, as patients who achieved pCR after NAST had a better long-term outcome compared with those who did not. In this paper, we attempt to prove our hypothesis. We performed a comprehensive analysis of the therapeutic effects of NAST and adjuvant systemic therapy (AST) in EBC to determine whether pCR, as a phenomenon that can only be achieved at NAST, improves prognosis per se. We used published papers as a source of data, which had a decisive influence on the formation of the modern attitude towards EBC therapy. We were unable to find any evidence supporting the use of pCR as a desired therapeutic goal because NAST (reinforced by pCR) was never demonstrated to be superior to AST in any context.

Effect of Longitudinal Variation in Tumor Volume Estimation for MRI-guided Personalization of Breast Cancer Neoadjuvant Treatment

Purpose To investigate the impact of longitudinal variation in functional tumor volume (FTV) underestimation and overestimation in predicting pathologic complete response (pCR) after neoadjuvant chemotherapy (NAC). Materials and Methods Women with breast cancer who were enrolled in the prospective I-SPY 2 TRIAL (Investigation of Serial Studies to Predict Your Therapeutic Response with Imaging and Molecular Analysis 2) from May 2010 to November 2016 were eligible for this retrospective analysis. Participants underwent four MRI examinations during NAC treatment. FTV was calculated based on automated segmentation. Baseline FTV before treatment (FTV0) and the percentage of FTV change at early treatment and inter-regimen time points relative to baseline (∆FTV1 and ∆FTV2, respectively) were classified into high-standard or standard groups based on visual assessment of FTV under- and overestimation. Logistic regression models predicting pCR using single predictors (FTV0, ∆FTV1, and ∆FTV2) and multiple predictors (all three) were developed using bootstrap resampling with out-of-sample data evaluation with the area under the receiver operating characteristic curve (AUC) independently in each group. Results This study included 432 women (mean age, 49.0 years ± 10.6 [SD]). In the FTV0 model, the high-standard and standard groups showed similar AUCs (0.61 vs 0.62). The high-standard group had a higher estimated AUC compared with the standard group in the ∆FTV1 (0.74 vs 0.63), ∆FTV2 (0.79 vs 0.62), and multiple predictor models (0.85 vs 0.64), with a statistically significant difference for the latter two models (P = .03 and P = .01, respectively). Conclusion The findings in this study suggest that longitudinal variation in FTV estimation needs to be considered when using early FTV change as an MRI-based criterion for breast cancer treatment personalization. Keywords: Breast, Cancer, Dynamic Contrast-enhanced, MRI, Tumor Response ClinicalTrials.gov registration no. NCT01042379 Supplemental material is available for this article. © RSNA, 2023 See also the commentary by Ram in this issue.

MRI-based Quantification of Intratumoral Heterogeneity for Predicting Treatment Response to Neoadjuvant Chemotherapy in Breast Cancer

Background Breast cancer is highly heterogeneous, resulting in different treatment responses to neoadjuvant chemotherapy (NAC) among patients. A noninvasive quantitative measure of intratumoral heterogeneity (ITH) may be valuable for predicting treatment response. Purpose To develop a quantitative measure of ITH on pretreatment MRI scans and test its performance for predicting pathologic complete response (pCR) after NAC in patients with breast cancer. Materials and Methods Pretreatment MRI scans were retrospectively acquired in patients with breast cancer who received NAC followed by surgery at multiple centers from January 2000 to September 2020. Conventional radiomics (hereafter, C-radiomics) and intratumoral ecological diversity features were extracted from the MRI scans, and output probabilities of imaging-based decision tree models were used to generate a C-radiomics score and ITH index. Multivariable logistic regression analysis was used to identify variables associated with pCR, and significant variables, including clinicopathologic variables, C-radiomics score, and ITH index, were combined into a predictive model for which performance was assessed using the area under the receiver operating characteristic curve (AUC). Results The training data set was comprised of 335 patients (median age, 48 years [IQR, 42-54 years]) from centers A and B, and 590, 280, and 384 patients (median age, 48 years [IQR, 41-55 years]) were included in the three external test data sets. Molecular subtype (odds ratio [OR] range, 4.76-8.39 [95% CI: 1.79, 24.21]; all P < .01), ITH index (OR, 30.05 [95% CI: 8.43, 122.64]; P < .001), and C-radiomics score (OR, 29.90 [95% CI: 12.04, 81.70]; P < .001) were independently associated with the odds of achieving pCR. The combined model showed good performance for predicting pCR to NAC in the training data set (AUC, 0.90) and external test data sets (AUC range, 0.83-0.87). Conclusion A model that combined an index created from pretreatment MRI-based imaging features quantitating ITH, C-radiomics score, and clinicopathologic variables showed good performance for predicting pCR to NAC in patients with breast cancer. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Rauch in this issue.

Early-Stage Triple-Negative Breast Cancer Journey: Beginning, End, and Everything in Between

Triple-negative breast cancer (TNBC) is a very heterogeneous and aggressive breast cancer subtype with a high risk of mortality, even if diagnosed early. The mainstay of early-stage breast cancer includes systemic chemotherapy and surgery, with or without radiation therapy. More recently, immunotherapy is approved to treat TNBC, but managing immune-rated adverse events while balancing efficacy is a challenge. The purpose of this review is to highlight the current treatment recommendations for early-stage TNBC and the management of immunotherapy toxicities.

Application of triple evaluation method in predicting the efficacy of neoadjuvant therapy for breast cancer

OBJECTIVE

To analyze the factors related to the efficacy of neoadjuvant therapy for breast cancer and find appropriate evaluation methods for evaluating the efficacy of neoadjuvant therapy METHODS: A total of 143 patients with breast cancer treated by neoadjuvant chemotherapy at Baotou Cancer Hospital were retrospectively analyzed. The chemotherapy regimen was mainly paclitaxel combined with carboplatin for 1 week, docetaxel combined with carboplatin for 3 weeks, and was replaced with epirubicin combined with cyclophosphamide after evaluation of disease progression. All HER2-positive patients were treated with simultaneous targeted therapy, including trastuzumab single-target therapy and trastuzumab combined with pertuzumab double-target therapy. Combined with physical examination, color Doppler ultrasound, and magnetic resonance imaging (MRI), a systematic evaluation system was initially established-the "triple evaluation method." A baseline evaluation was conducted before treatment. The efficacy was evaluated by physical examination and color Doppler every cycle, and the efficacy was evaluated by physical examination, color Doppler, and MRI every two cycles.

RESULTS

The increase in ultrasonic blood flow after treatment could affect the efficacy of monitoring. The presence of two preoperative time-signal intensity curves is a therapeutically effective protective factor for inflow. The triple evaluation determined by physical examination, color Doppler ultrasound, and MRI in determining clinical efficacy is consistent with the effectiveness of the pathological gold standard.

CONCLUSION

The therapeutic effect of neoadjuvant therapy can be better evaluated by combining clinical physical examination, color ultrasound, and nuclear magnetic resonance evaluation. The three methods complement each other to avoid the insufficient evaluation of a single method, which is convenient for most prefecty-level hospitals. Additionally, this method is simple, feasible, and suitable for promotion.

Four-Dimensional Machine Learning Radiomics for the Pretreatment Assessment of Breast Cancer Pathologic Complete Response to Neoadjuvant Chemotherapy in Dynamic Contrast-Enhanced MRI

BACKGROUND

Breast cancer response to neoadjuvant chemotherapy (NAC) is typically evaluated through the assessment of tumor size reduction after a few cycles of NAC. In case of treatment ineffectiveness, this results in the patient suffering potentially severe secondary effects without achieving any actual benefit.

PURPOSE

To identify patients achieving pathologic complete response (pCR) after NAC by spatio-temporal radiomic analysis of dynamic contrast-enhanced (DCE) MRI images acquired before treatment.

STUDY TYPE

Single-center, retrospective.

POPULATION

A total of 251 DCE-MRI pretreatment images of breast cancer patients.

FIELD STRENGTH/SEQUENCE

1.5 T/3 T, T1-weighted DCE-MRI.

ASSESSMENT

Tumor and peritumoral regions were segmented, and 348 radiomic features that quantify texture temporal variation, enhancement kinetics heterogeneity, and morphology were extracted. Based on subsets of features identified through forward selection, machine learning (ML) logistic regression models were trained separately with all images and stratifying on cancer molecular subtype and validated with leave-one-out cross-validation.

STATISTICAL TESTS

Feature significance was assessed using the Mann-Whitney U-test. Significance of the area under the receiver operating characteristics (ROC) curve (AUC) of the ML models was assessed using the associated 95% confidence interval (CI). Significance threshold was set to 0.05, adjusted with Bonferroni correction.

RESULTS

Nine features related to texture temporal variation and enhancement kinetics heterogeneity were significant in the discrimination of cases achieving pCR vs. non-pCR. The ML models achieved significant AUC of 0.707 (all cancers, n = 251, 59 pCR), 0.824 (luminal A, n = 107, 14 pCR), 0.823 (luminal B, n = 47, 15 pCR), 0.844 (HER2 enriched, n = 25, 11 pCR), 0.803 (triple negative, n = 72, 19 pCR).

DATA CONCLUSIONS

Differences in imaging phenotypes were found between complete and noncomplete responders. Furthermore, ML models trained per cancer subtype achieved high performance in classifying pCR vs. non-pCR cases. They may, therefore, have potential to help stratify patients according to the level of response predicted before treatment, pending further validation with larger prospective cohorts.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: Stage 4.

Differences Between Ipsilateral and Contralateral Early Parenchymal Enhancement Kinetics Predict Response of Breast Cancer to Neoadjuvant Therapy

RATIONALE AND OBJECTIVES

To determine whether kinetics measured with ultrafast dynamic contrast-enhanced magnetic resonance imaging in tumor and normal parenchyma pre- and post-neoadjuvant therapy (NAT) can predict the response of breast cancer to NAT.

MATERIALS AND METHODS

Twenty-four patients with histologically confirmed invasive breast cancer were enrolled. They were scanned with ultrafast dynamic contrast-enhanced magnetic resonance imaging (3-7 seconds/frame) pre- and post-NAT. Four kinetic parameters were calculated in the segmented tumors, and ipsi- and contra-lateral normal parenchyma: (1) tumor (tSE30) or background parenchymal relative enhancement at 30 seconds (BPE30), (2) maximum relative enhancement slope (MaxSlope), (3) bolus arrival time (BAT), and (4) area under relative signal enhancement curve for the initial 30 seconds (AUC30). The tumor kinetics and the differences between ipsi- and contra-lateral parenchymal kinetics were compared for patients achieving pathologic complete response (pCR) vs those who had residual disease after NAT. The chi-squared test and two-sided t-test were used for baseline demographics. The Wilcoxon rank sum test and one-way analysis of variance were used for differential responses to therapy.

RESULTS

Patients with similar pre-NAT mean BPE30, median BAT and mean AUC30 in the ipsi- and contralateral normal parenchyma were more likely to achieve pCR following NAT (p < 0.02). Patients classified as having residual cancer burden (RCB) II after NAT showed higher post-NAT tSE30 and tumor AUC30 and higher post-NAT MaxSlope in ipsilateral normal parenchyma compared to those classified as RCB I or pCR (p < 0.05).

CONCLUSION

Bilateral asymmetry in normal parenchyma could predict treatment outcome prior to NAT. Post-NAT tumor kinetics could evaluate the aggressiveness of residual tumor.

Dynamic characterization of breast cancer response to neoadjuvant therapy using biophysical metrics of spatial proliferation

Current tools to assess breast cancer response to neoadjuvant chemotherapy cannot reliably predict disease eradication, which if possible, could allow early cessation of therapy. In this work, we assessed the ability of an image data-driven mathematical modeling approach for dynamic characterization of breast cancer response to neoadjuvant therapy. We retrospectively analyzed patients enrolled in the I-SPY 2 TRIAL at the Atrium Health Wake Forest Baptist Comprehensive Cancer Center. Patients enrolled on the study received four MR imaging examinations during neoadjuvant therapy with acquisitions at baseline (T₀), 3-weeks/early-treatment (T₁), 12-weeks/mid-treatment (T₂), and completion of therapy prior to surgery (T₃). We use a biophysical mathematical model of tumor growth to generate spatial estimates of tumor proliferation to characterize the dynamics of treatment response. Using histogram summary metrics to quantify estimated tumor proliferation maps, we found strong correlation of mathematical model-estimated tumor proliferation with residual cancer burden, with Pearson correlation coefficients ranging from 0.88 and 0.97 between T₀ and T₂, representing a significant improvement from conventional assessment methods of change in mean apparent diffusion coefficient and functional tumor volume. This data shows the significant promise of imaging-based biophysical mathematical modeling methods for dynamic characterization of patient-specific response to neoadjuvant therapy with correlation to residual disease outcomes.

Deep learning in breast imaging

Millions of breast imaging exams are performed each year in an effort to reduce the morbidity and mortality of breast cancer. Breast imaging exams are performed for cancer screening, diagnostic work-up of suspicious findings, evaluating extent of disease in recently diagnosed breast cancer patients, and determining treatment response. Yet, the interpretation of breast imaging can be subjective, tedious, time-consuming, and prone to human error. Retrospective and small reader studies suggest that deep learning (DL) has great potential to perform medical imaging tasks at or above human-level performance, and may be used to automate aspects of the breast cancer screening process, improve cancer detection rates, decrease unnecessary callbacks and biopsies, optimize patient risk assessment, and open up new possibilities for disease prognostication. Prospective trials are urgently needed to validate these proposed tools, paving the way for real-world clinical use. New regulatory frameworks must also be developed to address the unique ethical, medicolegal, and quality control issues that DL algorithms present. In this article, we review the basics of DL, describe recent DL breast imaging applications including cancer detection and risk prediction, and discuss the challenges and future directions of artificial intelligence-based systems in the field of breast cancer.

Assessment of Cone-Beam Breast Computed Tomography for Predicting Pathologic Response to Neoadjuvant Chemotherapy in Breast Cancer: A Prospective Study

Background

Response surveillance of neoadjuvant chemotherapy is needed to facilitate treatment decisions. We aimed to assess the imaging features of cone-beam breast computed tomography (CBBCT) for predicting the pathologic response of breast cancer after neoadjuvant chemotherapy.

Methods

This prospective study included 81 women with locally advanced breast cancer who underwent neoadjuvant chemotherapy from August 2017 to January 2021. All patients underwent CBBCT before treatment, and 55 and 65 patients underwent CT examinations during the midtreatment (3 cycles) and late-treatment phases (7 cycles), respectively. Clinical information and quantitative parameters such as the diameter, volume, surface area, and CT density were compared between pathologic responders and nonresponders using the T–test and the Mann–Whitney U test. The performance of meaningful parameters was evaluated with the receiver operating characteristic curve, sensitivity, and specificity.

Results

The quantitative results for the segmented volume, segmented surface area, segmented volume reduction, maximum enhancement ratio, wash-in rate and two-minute enhancement value in the mid- and late-treatment periods had predictive value for pathologic complete response. The area under the curve for the prediction model after multivariate regression analysis was 0.874.

Conclusion

After comparing the outcomes of each timepoint, mid- and late-treatment parameters can be used to predict pathologic outcome. The late-treatment parameters showed significant value with a predictive model.

Quantitative Predictors of Response to Neoadjuvant Chemotherapy on Dynamic Contrast-enhanced 3T Breast MRI

OBJECTIVE

To assess whether changes in quantitative parameters on breast MRI better predict pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast cancer than change in volume.

METHODS

This IRB-approved retrospective study included women with newly diagnosed breast cancer who underwent 3T MRI before and during NAC from January 2013 to December 2019 and underwent surgery at our institution. Clinical data such as age, histologic diagnosis and grade, biomarker status, clinical stage, maximum index cancer dimension and volume, and surgical pathology (presence or absence of in-breast pCR) were collected. Quantitative parameters were calculated using software. Correlations between clinical features and MRI quantitative measures in pCR and non-pCR groups were assessed using univariate and multivariate logistic regression.

RESULTS

A total of 182 women with a mean age of 52 years (range, 26-79 years) and 187 cancers were included. Approximately 45% (85/182) of women had pCR at surgery. Stepwise multivariate regression analysis showed statistical significance for changes in quantitative parameters (increase in time to peak and decreases in peak enhancement, wash out, and Kep [efflux rate constant]) for predicting pCR. These variables in combination predicted pCR with 81.2% accuracy and an area under the curve (AUC) of 0.878. The AUCs of change in index cancer volume and maximum dimension were 0.767 and 0.613, respectively.

CONCLUSION

Absolute changes in quantitative MRI parameters between pre-NAC MRI and intra-NAC MRI could help predict pCR with excellent accuracy, which was greater than changes in index cancer volume and maximum dimension.

Post-Processing Bias Field Inhomogeneity Correction for Assessing Background Parenchymal Enhancement on Breast MRI as a Quantitative Marker of Treatment Response

Background parenchymal enhancement (BPE) of breast fibroglandular tissue (FGT) in dynamic contrast-enhanced breast magnetic resonance imaging (MRI) has shown an association with response to neoadjuvant chemotherapy (NAC) in patients with breast cancer. Fully automated segmentation of FGT for BPE calculation is a challenge when image artifacts are present. Low spatial frequency intensity nonuniformity due to coil sensitivity variations is known as bias or inhomogeneity and can affect FGT segmentation and subsequent BPE measurement. In this study, we utilized the N4ITK algorithm for bias correction over a restricted bilateral breast volume and compared the contralateral FGT segmentations based on uncorrected and bias-corrected images in three MRI examinations at pre-treatment, early treatment and inter-regimen timepoints during NAC. A retrospective analysis of 2 cohorts was performed: one with 735 patients enrolled in the multi-center I-SPY 2 TRIAL and the sub-cohort of 340 patients meeting a high-quality benchmark for segmentation. Bias correction substantially increased the FGT segmentation quality for 6.3–8.0% of examinations, while it substantially decreased the quality for no examination. Our results showed improvement in segmentation quality and a small but statistically significant increase in the resulting BPE measurement after bias correction at all timepoints in both cohorts. Continuing studies are examining the effects on pCR prediction.

Neoadjuvant approach in patients with early breast cancer: patient assessment, staging, and planning

Neoadjuvant treatment (NAT) has become an option in early stage (stage I-II) breast cancer (EBC). New advances in systemic and targeted therapies have increased rates of pathologic complete response increasing the number of patients undergoing NAT. Clear benefits of NAT are downstaging the tumor and the axillary nodes to de-escalate surgery and to evaluate response to treatment. Selection of patients for NAT in EBC rely in several factors that are related to patient characteristics (i.e, age and comorbidities), to tumor histology, to stage at diagnosis and to the potential changes in surgical or adjuvant treatments when NAT is administered.

Imaging and histologic confirmation is performed to assess extent of disease y to confirm diagnosis. Besides mammogram and ultrasound, functional breast imaging MRI has been incorporated to better predict treatment response and residual disease. Contrast enhanced mammogram (CEM), shear wave elastography (SWE), or Dynamic Optical Breast Imaging (DOBI) are emerging techniques under investigation for assessment of response to neoadjuvant therapy as well as for predicting response. Surgical plan should be delineated after NAT taking into account baseline characteristics, tumor response and patient desire.

In the COVID era, we have witnessed also the increasing use of NAT in patients who may be directed to surgery, unable to have it performed as surgery has been reserved for emergency cases only.

An MRI-Based Radiomics Model for Predicting the Benignity and Malignancy of BI-RADS 4 Breast Lesions

Objectives

The probability of Breast Imaging Reporting and Data Systems (BI-RADS) 4 lesions being malignant is 2%–95%, which shows the difficulty to make a diagnosis. Radiomics models based on magnetic resonance imaging (MRI) can replace clinicopathological diagnosis with high performance. In the present study, we developed and tested a radiomics model based on MRI images that can predict the malignancy of BI-RADS 4 breast lesions.

Methods

We retrospective enrolled a total of 216 BI-RADS 4 patients MRI and clinical information. We extracted 3,474 radiomics features from dynamic contrast-enhanced (DCE), T₂-weighted images (T₂WI), and diffusion-weighted imaging (DWI) MRI images. Least absolute shrinkage and selection operator (LASSO) and logistic regression were used to select features and build radiomics models based on different sequence combinations. We built eight radiomics models which were based on DCE, DWI, T₂WI, DCE+DWI, DCE+T₂WI, DWI+T₂WI, and DCE+DWI+T₂WI and a clinical predictive model built based on the visual assessment of radiologists. A nomogram was constructed with the best radiomics signature combined with patient characteristics. The calibration curves for the radiomics signature and nomogram were conducted, combined with the Hosmer-Lemeshow test.

Results

Pearson’s correlation was used to eliminate 3,329 irrelevant features, and then LASSO and logistic regression were used to screen the remaining feature coefficients for each model we built. Finally, 12 related features were obtained in the model which had the best performance. These 12 features were used to build a radiomics model in combination with the actual clinical diagnosis of benign or malignant lesion labels we have obtained. The best model built by 12 features from the 3 sequences has an AUC value of 0.939 (95% CI, 0.884-0.994) and an accuracy of 0.931 in the testing cohort. The sensitivity, specificity, precision and Matthews correlation coefficient (MCC) of testing cohort are 0.932, 0.923, 0.982, and 0.791, respectively. The nomogram has also been verified to have calibration curves with good overlap.

Conclusions

Radiomics is beneficial in the malignancy prediction of BI-RADS 4 breast lesions. The radiomics predictive model built by the combination of DCE, DWI, and T₂WI sequences has great application potential.

Circulating tumor DNA and magnetic resonance imaging to predict neoadjuvant chemotherapy response and recurrence risk

We investigated whether serial measurements of circulating tumor DNA (ctDNA) and functional tumor volume (FTV) by magnetic resonance imaging (MRI) can be combined to improve prediction of pathologic complete response (pCR) and estimation of recurrence risk in early breast cancer patients treated with neoadjuvant chemotherapy (NAC). We examined correlations between ctDNA and FTV, evaluated the additive value of ctDNA to FTV-based predictors of pCR using area under the curve (AUC) analysis, and analyzed the impact of FTV and ctDNA on distant recurrence-free survival (DRFS) using Cox regressions. The levels of ctDNA (mean tumor molecules/mL plasma) were significantly correlated with FTV at all time points (p < 0.05). Median FTV in ctDNA-positive patients was significantly higher compared to those who were ctDNA-negative (p < 0.05). FTV and ctDNA trajectories in individual patients showed a general decrease during NAC. Exploratory analysis showed that adding ctDNA information early during treatment to FTV-based predictors resulted in numerical but not statistically significant improvements in performance for pCR prediction (e.g., AUC 0.59 vs. 0.69, p = 0.25). In contrast, ctDNA-positivity after NAC provided significant additive value to FTV in identifying patients with increased risk of metastatic recurrence and death (p = 0.004). In this pilot study, we demonstrate that ctDNA and FTV were correlated measures of tumor burden. Our preliminary findings based on a limited cohort suggest that ctDNA at surgery improves FTV as a predictor of metastatic recurrence and death. Validation in larger studies is warranted.

The Way of the Future: Personalizing Treatment Plans Through Technology

Advances in tissue analysis methods, image analysis, high-throughput molecular profiling, and computational tools increasingly allow us to capture and quantify patient-to patient variations that impact cancer risk, prognosis, and treatment response. Statistical models that integrate patient-specific information from multiple sources (e.g., family history, demographics, germline variants, imaging features) can provide individualized cancer risk predictions that can guide screening and prevention strategies. The precision, quality, and standardization of diagnostic imaging are improving through computer-aided solutions, and multigene prognostic and predictive tests improved predictions of prognosis and treatment response in various cancer types. A common theme across many of these advances is that individually moderately informative variables are combined into more accurate multivariable prediction models. Advances in machine learning and the availability of large data sets fuel rapid progress in this field. Molecular dissection of the cancer genome has become a reality in the clinic, and molecular target profiling is now routinely used to select patients for various targeted therapies. These technology-driven increasingly more precise and quantitative estimates of benefit versus risk from a given intervention empower patients and physicians to tailor treatment strategies that match patient values and expectations.

Circulating tumor DNA in neoadjuvant-treated breast cancer reflects response and survival

BACKGROUND

Pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) is strongly associated with favorable outcome. We examined the utility of serial circulating tumor DNA (ctDNA) testing for predicting pCR and risk of metastatic recurrence.

PATIENTS AND METHODS

Cell-free DNA (cfDNA) was isolated from 291 plasma samples of 84 high-risk early breast cancer patients treated in the neoadjuvant I-SPY 2 TRIAL with standard NAC alone or combined with MK-2206 (AKT inhibitor) treatment. Blood was collected at pretreatment (T0), 3 weeks after initiation of paclitaxel (T1), between paclitaxel and anthracycline regimens (T2), or prior to surgery (T3). A personalized ctDNA test was designed to detect up to 16 patient-specific mutations (from whole-exome sequencing of pretreatment tumor) in cfDNA by ultra-deep sequencing. The median follow-up time for survival analysis was 4.8 years.

RESULTS

At T0, 61 of 84 (73%) patients were ctDNA positive, which decreased over time (T1: 35%; T2: 14%; and T3: 9%). Patients who remained ctDNA positive at T1 were significantly more likely to have residual disease after NAC (83% non-pCR) compared with those who cleared ctDNA (52% non-pCR; odds ratio 4.33, P = 0.012). After NAC, all patients who achieved pCR were ctDNA negative (n = 17, 100%). For those who did not achieve pCR (n = 43), ctDNA-positive patients (14%) had a significantly increased risk of metastatic recurrence [hazard ratio (HR) 10.4; 95% confidence interval (CI) 2.3-46.6]; interestingly, patients who did not achieve pCR but were ctDNA negative (86%) had excellent outcome, similar to those who achieved pCR (HR 1.4; 95% CI 0.15-13.5).

CONCLUSIONS

Lack of ctDNA clearance was a significant predictor of poor response and metastatic recurrence, while clearance was associated with improved survival even in patients who did not achieve pCR. Personalized monitoring of ctDNA during NAC of high-risk early breast cancer may aid in real-time assessment of treatment response and help fine-tune pCR as a surrogate endpoint of survival.