Sample size determination for prediction models via learning-type curves.

This article is concerned with sample size determination methodology for prediction models. We propose to combine the individual calculations via learning-type curves. We suggest two distinct ways of doing so, a deterministic skeleton of a learning curve and a Gaussian process centered upon its deterministic counterpart. We employ several learning algorithms for modeling the primary endpoint and distinct measures for trial efficacy. We find that the performance may vary with the sample size, but borrowing information across sample size universally improves the performance of such calculations. The Gaussian process-based learning curve appears more robust and statistically efficient, while computational efficiency is comparable. We suggest that anchoring against historical evidence when extrapolating sample sizes should be adopted when such data are available. The methods are illustrated on binary and survival endpoints.

The PARTNER trial of neoadjuvant olaparib with chemotherapy in triple-negative breast cancer.

PARTNER is a prospective, phase II-III, randomized controlled clinical trial that recruited patients with triple-negative breast cancer1,2, who were germline BRCA1 and BRCA2 wild type3. Here we report the results of the trial. Patients (n = 559) were randomized on a 1:1 basis to receive neoadjuvant carboplatin-paclitaxel with or without 150 mg olaparib twice daily, on days 3 to 14, of each of four cycles (gap schedule olaparib, research arm) followed by three cycles of anthracycline-based chemotherapy before surgery. The primary end point was pathologic complete response (pCR)4, and secondary end points included event-free survival (EFS) and overall survival (OS)5. pCR was achieved in 51% of patients in the research arm and 52% in the control arm (P = 0.753). Estimated EFS at 36 months in the research and control arms was 80% and 79% (log-rank P > 0.9), respectively; OS was 90% and 87.2% (log-rank P = 0.8), respectively. In patients with pCR, estimated EFS at 36 months was 90%, and in those with non-pCR it was 70% (log-rank P < 0.001), and OS was 96% and 83% (log-rank P < 0.001), respectively. Neoadjuvant olaparib did not improve pCR rates, EFS or OS when added to carboplatin-paclitaxel and anthracycline-based chemotherapy in patients with triple-negative breast cancer who were germline BRCA1 and BRCA2 wild type. ClinicalTrials.gov ID: NCT03150576 .

Playing by the Rules? Tracking U.S. Hospitals' Responses to Federal Price Transparency Regulation.

GOAL: As of January 1, 2021, the Centers for Medicare & Medicaid Services requires most U.S. hospitals to publish pricing information on their website to help consumers make decisions regarding services and to transform negotiations with health insurers. For this study, we evaluated changes in hospitals' compliance with the federal price transparency rule after the first year of enactment, during which the Centers for Medicare & Medicaid Services increased the penalty for noncompliance. METHODS: Using a nationally representative random sample of 470 hospitals, we assessed compliance with both parts of the hospital transparency rule (publishing a machine-readable price database and a consumer shopping tool) in the first quarter of 2022 and compared its baseline level in the first quarter of 2021. Using data from the American Hospital Association and Clarivate, we next assessed how compliance varied by hospital factors (ownership, number of beds, system membership, teaching status, type of electronic health record system), market factors (hospital and insurer market concentration), and the estimated change in penalty for noncompliance. PRINCIPAL FINDINGS: By early 2022, 46% of hospitals had posted both machine-readable and consumer-shoppable data, an increase of 24% from the prior year. Almost 9 in 10 hospitals had complied with the consumer-shoppable data requirement by early 2022. Larger hospitals and public hospitals had lower probabilities of baseline compliance with the machine-readable and consumer-shoppable requirements, respectively, although public hospitals were significantly more likely to become compliant with the consumer-shoppable requirement by 2022. Higher hospital market concentration was also associated with higher baseline compliance for both the machine-readable and consumer-shoppable requirements. Furthermore, our analyses found that hospitals with certain electronic health record systems were more likely to comply with the consumer-shoppable requirement in 2021 and became increasingly compliant with the machine-readable requirement in 2022. Finally, we found that hospitals with a larger estimated penalty were more likely to become compliant with the machine-readable requirement. PRACTICAL APPLICATIONS: Longitudinal analyses of compliance with the federal price transparency rule are valuable for monitoring changes in hospitals' behavior and assessing whether compliance changes vary systematically for specific types of hospitals and/or market structures. Our results suggest a trend toward increased hospital compliance between 2021 and 2022. Although hospitals perceive the consumer-shopping tools as being the most impactful, the value of this information depends on whether it is comprehensible and comparable across hospitals. The new price transparency rule has facilitated the creation of new data that have the potential to significantly alter the competitive landscape for hospitals and may require hospital leaders to consider how their organizational strategies change concerning their engagement with payers and patients. Finally, greater price transparency is likely to bolster national policy discussions related to price variation, affordability, and the role of regulation in healthcare markets.

To be or not to be compliant? Hospitals' initial strategic responses to the federal price transparency rule.

OBJECTIVE: To understand US hospitals' initial strategic responses to the federal price transparency rule that took effect January 2021. DATA SOURCES AND STUDY SETTING: Primary interview data collected from 12 not-for-profit hospital organizations in six US metropolitan markets. All but one organization were multihospital systems; the 12 organizations represent a total of 81 hospitals. STUDY DESIGN: Exploratory, cross-sectional, qualitative interview study of a convenience sample of hospital organizations across six geographically and compliance diverse markets. DATA COLLECTION/EXTRACTION METHODS: In-depth, semi-structured, qualitative interviews with 16 key informants across sampled organizations between November 2021 and March 2022. Interviews solicited data about internal organizational factors and external market factors affecting strategic responses. Transcribed interviews were de-identified, coded, and analyzed using the constant comparative method. PRINCIPAL FINDINGS: Hospitals' strategic responses were influenced internally by the degree of the regulation's alignment with organizational values and goals, and task complexity vis-a-vis available resources. We found extensive variation in organizational capabilities to comply, and all but one organization relied on consultants and vendors to some degree. Key external factors driving strategic responses were hospitals' variable perceptions about how available price information would affect their competitive position, bottom line, and reputation. Organizations with more confidence in their interpretation of the environment, including how peers or purchasers would behave, and greater clarity in their own organization's position and goals, had more definitive initial strategic responses. In the first year, organizations' strategic responses skewed toward compliance, especially for the rule's consumer shopping requirements. CONCLUSIONS: A deeper understanding of the realities of operationalizing price transparency policy for hospitals is needed to improve its impact.

Do minimum wage laws affect employer-sponsored insurance provision?

Employers may respond to minimum wage increases by adjusting their health benefits. We examine the impact of state minimum wage increases on employer health benefit offerings using the 2002-2020 Medical Expenditure Panel Survey - Insurance/Employer Component data. Our primary regression specifications are difference-in-differences models that estimate the relationship between within-state changes in employer-sponsored insurance and minimum wage laws over time. We find that a $1 increase in minimum wages is associated with a 0.92 percentage point (p.p.) decrease in the percentage of employers offering health insurance, largely driven by small employers and employers with a greater share of low-wage employees. A $1 increase is also associated with a 1.83 p.p. increase in the prevalence of plans with a deductible requirement, but we do not find consistent evidence that other benefit characteristics are affected. We find no consequent change in uninsurance, likely explained by an increase in Medicaid enrollment.

Clinical validation of an elastin-derived trifunctional peptide for skin regeneration.

Aging is associated with progressive skin fragility, characterized in part by extracellular matrix (ECM) fragmentation. This degradation produces matrikines which have an impact on ECM rremodeling. Our group previously designed and characterized a trifunctional peptide (TFP), constituted of i) an elastokine motif (VGVAPG)3, able to increase the expression of matrix constituent through the stimulation of the elastin-binding protein receptor, ii) a tripeptide inhibiting matrix metalloproteinase-1 activity (GIL), and iii) a linker domain acting as a competitive substrate for urokinase (RVRL). TFP was shown to activate the production of matrix constituents while inhibiting Matrix MetalloProtease MMP-1 in vitro on fibroblasts and ex vivo on skin explants. OBJECTIVE: In the present study, TFP properties were evaluated in a clinical assay. METHODS: Twenty-two volunteers applied a TFP-based cream on one hemi-face and a placebo-based cream on the other hemi-face, twice a day during 28 days, before undergoing a surgical lifting. Cutometry and skin relief measurements were performed at days 0 and 28, and skin explants from lifting surgery were used for histological analyses. RESULTS: Cutometry and skin relief measurements reveal TFP firming properties and wrinkle depth decrease in 28 days on TFP- as compared to placebo-treated hemi-faces. These results are confirmed by histological analyses showing an increase of the ratio between basal lamina and stratum corneum. Furthermore, immunostaining of collagen reveals a modification of the ratio between type I and III collagens. CONCLUSION: The combined analysis of phenotypic and histologic parameters demonstrates a reorganization of the ECM towards a regenerative profile upon TFP treatment.

Was Unstable Medicaid Coverage Among Older Medicare Beneficiaries Associated With Worse Clinical Outcomes? Evidence From the Delivery of Breast Cancer Care.

BACKGROUND: Medicare and Medicaid dually eligible beneficiaries (duals) could experience Medicaid coverage changes without losing Medicaid. It is unknown whether health care use and clinical outcomes among elderly duals with coverage changes would be like those among duals without coverage changes or duals ever lost Medicaid and whether various types of unstable coverage due to income/asset changes are associated with worse clinical outcomes. OBJECTIVES: Examine the associations of unstable Medicaid coverage with clinical outcomes among older Medicare beneficiaries. RESEARCH DESIGN: Population-based cohort study. SUBJECTS: A total of 131,202 women newly diagnosed with breast cancer at 65 years and older between 2007 and 2015 were identified from the Surveillance, Epidemiology, and End Results-Medicare linked database. MEASURES: We examined 2 types of unstable Medicaid coverage: (1) those who had changes in the types of Medicaid support they received and (2) those who ever lost Medicaid. We examined outcomes that predict better cancer survival and involve the use of inpatient and outpatient services and prescription drugs: early diagnosis, receiving surgery, receiving radiation, hormonal therapy adherence, and discontinuation. We used logistic regressions to estimate the predicted probabilities of outcomes for dual groups. RESULTS: Duals had poorer outcomes than those who were "never dual." Women with the 2 types of unstable Medicaid coverage had similarly worse outcomes than those with stable coverage. Those with stable coverage had similar outcomes regardless of the generosity of Medicaid support. CONCLUSIONS: These patterns are concerning and, in the context of well-defined clinical guidelines for beneficial treatments that extend survival, point to the importance of stable insurance coverage and income.

Comparison of tumor-informed and tumor-naïve sequencing assays for ctDNA detection in breast cancer.

Analysis of circulating tumor DNA (ctDNA) to monitor cancer dynamics and detect minimal residual disease has been an area of increasing interest. Multiple methods have been proposed but few studies have compared the performance of different approaches. Here, we compare detection of ctDNA in serial plasma samples from patients with breast cancer using different tumor-informed and tumor-naïve assays designed to detect structural variants (SVs), single nucleotide variants (SNVs), and/or somatic copy-number aberrations, by multiplex PCR, hybrid capture, and different depths of whole-genome sequencing. Our results demonstrate that the ctDNA dynamics and allele fractions (AFs) were highly concordant when analyzing the same patient samples using different assays. Tumor-informed assays showed the highest sensitivity for detection of ctDNA at low concentrations. Hybrid capture sequencing targeting between 1,347 and 7,491 tumor-identified mutations at high depth was the most sensitive assay, detecting ctDNA down to an AF of 0.00024% (2.4 parts per million, ppm). Multiplex PCR targeting 21-47 tumor-identified SVs per patient detected ctDNA down to 0.00047% AF (4.7 ppm) and has potential as a clinical assay.

Evaluating the association between expanded coverage of direct-to-consumer telemedicine and downstream utilization and quality of care for urinary tract infections and sinusitis.

OBJECTIVE: To compare direct-to-consumer (DTC) telemedicine and in-person visits in rates of testing, follow-up health care use, and quality for urinary tract infections (UTIs) and sinusitis. DATA SOURCE: The Minnesota All Payer Claims Data provided 2008-2015 administrative claims data. STUDY DESIGN: Using a difference-in-differences approach, we compared episodes of care for UTIs and sinusitis among enrollees of health plans introducing coverage for DTC telemedicine relative to those without DTC telemedicine coverage. Primary outcomes included number of laboratory tests, antibiotics filled, office and outpatient visits, emergency department (ED) visits, and standardized spending, based on standardized prices of health services. DATA COLLECTION: The study sample included non-elderly enrollees of commercial health insurance plans. We constructed 30-day episodes of care initiated by a DTC telemedicine or in-person visit. PRINCIPAL FINDINGS: The UTI and sinusitis samples were comprised of 215,134 and 624,630 episodes of care, respectively. Following the introduction of coverage for DTC telemedicine, 15.7% of UTI episodes and 8.9% of sinusitis episodes were initiated with DTC telemedicine. Compared to episodes without coverage for DTC telemedicine, UTI episodes with coverage had 0.25 fewer lab tests (95% CI: -0.33, -0.18; p < 0.001), lower standardized spending for the first UTI visit (-$11.18 [95% CI: -$21.62, -$0.75]; p < 0.05), and no change in office and outpatient visits, ED visits, antibiotics filled, or standardized medical spending. Sinusitis episodes with coverage for DTC telemedicine had fewer antibiotics filled (-0.08 [95% CI: -0.14, -0.01]; p < 0.05) and a very small increase in ED visits (0.001 [95% CI: 0.001, 0.010]; p < 0.05), but no change in lab tests, office and outpatient visits, or standardized medical spending. CONCLUSIONS: Among commercially insured patients, coverage of DTC telemedicine was associated with reductions in antibiotics for sinusitis and laboratory tests for UTI without changes in downstream total office and outpatient visits or changes in ED visits.

Mainstreaming of genomics in oncology: a nationwide survey of the genomics training needs of UK oncologists.

OBJECTIVE: Genomics is rapidly changing treatment paradigms for cancers, obligating oncologists to have good genomics knowledge. Through this survey, we aimed to assess the current understanding of cancer genomics among UK oncologists. METHODS: We conducted a web-based nation-wide self-assessment survey of the cancer genomics knowledge of UK clinical and medical oncology trainees and consultants. RESULTS: In total, 150 oncologists (81 consultants and 69 trainees) responded, representing 10% of UK oncologists.Formal training in genomics had not been received by 38.7% of oncologists and 92.7% identified a need for additional genomics training.In total, 71.3% self-reported to have good knowledge of defining somatic and germline mutations, falling to 35.3% for understanding principles of gene expression and regulation. Knowledge of cancer-predisposing syndromes was highest for Lynch syndrome (40.7% good knowledge) and lowest for multiple endocrine neoplasia (14.0% good knowledge).Overall, 49.0% of respondents had consented patients for germline testing, but 80.7% reported a lack of training in genetic counselling. CONCLUSION: Large knowledge gaps have been identified through this survey, highlighting the need for incorporation of improved formal training in cancer genomics for consultants and trainees, with an aim to equip oncologists for advances in clinical practice and to take up genetic mainstreaming confidently.

Precision Breast Cancer Medicine: Early Stage Triple Negative Breast Cancer-A Review of Molecular Characterisation, Therapeutic Targets and Future Trends.

Personalised approaches to the management of all solid tumours are increasing rapidly, along with wider accessibility for clinicians. Advances in tumour characterisation and targeted therapies have placed triple-negative breast cancers (TNBC) at the forefront of this approach. TNBC is a highly heterogeneous disease with various histopathological features and is driven by distinct molecular alterations. The ability to tailor individualised and effective treatments for each patient is of particular importance in this group due to the high risk of distant recurrence and death. The mainstay of treatment across all subtypes of TNBC has historically been cytotoxic chemotherapy, which is often associated with off-target tissue toxicity and drug resistance. Neoadjuvant chemotherapy is commonly used as it allows close monitoring of early treatment response and provides valuable prognostic information. Patients who achieve a complete pathological response after neoadjuvant chemotherapy are known to have significantly improved long-term outcomes. Conversely, poor responders face a higher risk of relapse and death. The identification of those subgroups that are more likely to benefit from breakthroughs in the personalised approach is a challenge of the current era where several targeted therapies are available. This review presents an overview of contemporary practice, and promising future trends in the management of early TNBC. Platinum chemotherapy, DNA damage response (DDR) inhibitors, immune checkpoint inhibitors, inhibitors of the PI3K-AKT-mTOR, and androgen receptor (AR) pathways are some of the increasingly studied therapies which will be reviewed. We will also discuss the growing evidence for less-developed agents and predictive biomarkers that are likely to contribute to the forthcoming advances in this field. Finally, we will propose a framework for the personalised management of TNBC based upon the integration of clinico-pathological and molecular features to ensure that long-term outcomes are optimised.

A game changer for bipolar disorder diagnosis using RNA editing-based biomarkers.

In clinical practice, differentiating Bipolar Disorder (BD) from unipolar depression is a challenge due to the depressive symptoms, which are the core presentations of both disorders. This misdiagnosis during depressive episodes results in a delay in proper treatment and a poor management of their condition. In a first step, using A-to-I RNA editome analysis, we discovered 646 variants (366 genes) differentially edited between depressed patients and healthy volunteers in a discovery cohort of 57 participants. After using stringent criteria and biological pathway analysis, candidate biomarkers from 8 genes were singled out and tested in a validation cohort of 410 participants. Combining the selected biomarkers with a machine learning approach achieved to discriminate depressed patients (n = 267) versus controls (n = 143) with an AUC of 0.930 (CI 95% [0.879-0.982]), a sensitivity of 84.0% and a specificity of 87.1%. In a second step by selecting among the depressed patients those with unipolar depression (n = 160) or BD (n = 95), we identified a combination of 6 biomarkers which allowed a differential diagnosis of bipolar disorder with an AUC of 0.935 and high specificity (Sp = 84.6%) and sensitivity (Se = 90.9%). The association of RNA editing variants modifications with depression subtypes and the use of artificial intelligence allowed developing a new tool to identify, among depressed patients, those suffering from BD. This test will help to reduce the misdiagnosis delay of bipolar patients, leading to an earlier implementation of a proper treatment.

Hormonal Therapy Drug Switching, Out-of-Pocket Costs, and Adherence Among Older Women With Breast Cancer.

BACKGROUND: Adherence to aromatase inhibitors (AIs) and tamoxifen has considerable survival benefits for postmenopausal women diagnosed with hormone receptor-positive breast cancer. Reduced out-of-pocket costs and treatment-related side effects could increase therapy adherence. Given that individuals' side effect profiles could differ across AIs, generic AI entry could facilitate switching between AIs to manage side effects and improve adherence. METHODS: From Surveillance, Epidemiology, and End Results-Medicare, we selected women first diagnosed with hormone receptor-positive breast cancer at age 65+ years and initiated an AI within 1 year of diagnosis between January 1, 2007, and May 31, 2008, or June 1, 2011, and December 31, 2012, and followed them for up to 2 years (N = 20 677). We estimated changes in probabilities of adherence with and without switching for Part D enrollees with and without the low-income subsidy (LIS vs non-LIS) before and after generic entry using linear probability models. Tests of statistical significance are 2-sided. RESULTS: After generic entry reduced out-of-pocket costs of AIs (larger reduction for non-LIS), the percentage of women who ever switched from one AI to another AI increased from 8.8% to 14.6% for non-LIS and from 7.3% to 12.5% for LIS. Adherence without switching increased by 8.0 percentage points (pp) for non-LIS (P < .001) but decreased by 4.9 pp (P < .001) for LIS. Adherence with switching increased for both non-LIS (6.4 pp, P < .001) and LIS (4.4 pp, P < .001). CONCLUSIONS: Increased switching after generic entry contributed to increased adherence, suggesting switching allowed better management of treatment-related side effects. Subsidized women also experienced increased adherence with switching after generic entry, suggesting that patients and physicians might not understand Part D benefit design when making decisions.

Association Between Provider-Sponsored Health Plan Ownership and Health Insurance Marketplace Plan Quality.

BACKGROUND: Over 15.3 million Americans relied on the individual health insurance market for health coverage in 2021. Yet, little is known about the relationships between the organizational characteristics of individual market health insurers and quality of coverage, particularly with respect to clinical outcomes. OBJECTIVE: To examine variation in marketplace insurers' quality performance and investigate how performance varies by insurer organizational characteristics. DESIGN: Retrospective cohort study. PARTICIPANTS: 381 insurer products, representing 184 unique insurers in 50 states in 2019 and 2020. MAIN MEASURES: Marketplace plan clinical quality measures reported in the 2019-2020 CMS Plan Quality Rating System dataset and insurer-product organizational attributes identified from several data sources, including non-profit ownership, Blue Cross Blue Shield Association membership, Medicaid focus and whether or not the insurer product is vertically integrated with a provider organization. KEY RESULTS: Among the 381 insurer products in this study, 35% are part of a provider-sponsored health plan (PSHP) and 70% of these entities received four stars or above for overall quality performance. Overall, PSHPs exhibited higher quality than non-PSHPs for both clinical quality management (0.36 increased on a 5-point scale; 95% CI = 0.11 to 0.62; P = 0.005) and enrollee experience (0.27; 95% CI = 0.03 to 0.50; P = 0.03) summary indicators. Medicaid focused insurers were associated with lower performance on enrollee experience, plan administration, and various outcomes related to clinical quality. CONCLUSIONS: Provider-sponsored health plans in the health insurance marketplaces are associated with higher-quality care, as measured by CMS clinical quality measures.

Association of neutrophil extracellular traps with the production of circulating DNA in patients with colorectal cancer.

We postulate that a significant part of circulating DNA (cirDNA) originates in the degradation of neutrophil extracellular traps (NETs). In this study, we examined the plasma level of two markers of NETs (myeloperoxidase (MPO) and neutrophil elastase (NE)), as well as cirDNA levels in 219 patients with a metastatic colorectal cancer (mCRC), and in 114 healthy individuals (HI). We found that in patients with mCRC the content of these analytes was (i) highly correlated, and (ii) all statistically different (p < 0.0001) than in HI (N = 114). These three NETs markers may readily distinguish between patients with mCRC from HI, (0.88, 0.86, 0.84, and 0.95 AUC values for NE, MPO, cirDNA, and NE + MPO + cirDNA, respectively). Concomitant analysis of anti-phospholipid (anti-cardiolipin), NE, MPO, and cirDNA plasma concentrations in patients with mCRC might have value for thrombosis prevention, and suggested that NETosis may be a critical factor in the immunological response/phenomena linked to tumor progression.

Taking the Pulse of Hospitals' Response to the New Price Transparency Rule.

As of January 1, 2021, most U.S. hospitals are required to publish pricing information on their website to promote more informed decision making by consumers regarding their care. In a nationally representative sample of 470 hospitals, we analyzed whether hospitals met price transparency information reporting requirements and the extent to which complete reporting was associated with ownership status, bed size category, system affiliation, and location in a metropolitan area. Fewer than one quarter of sampled hospitals met the price transparency information requirements of the new rule, which include five types of standard charges in machine-readable form and the consumer-shoppable display of 300 shoppable services. Our analyses of hospital reporting by organizational and market attributes revealed limited differences, with some exceptions for nonprofit and system-member hospitals demonstrating greater responsiveness with respect to the consumer-shoppable aspects of the rule.

Multi-omic machine learning predictor of breast cancer therapy response.

Breast cancers are complex ecosystems of malignant cells and the tumour microenvironment1. The composition of these tumour ecosystems and interactions within them contribute to responses to cytotoxic therapy2. Efforts to build response predictors have not incorporated this knowledge. We collected clinical, digital pathology, genomic and transcriptomic profiles of pre-treatment biopsies of breast tumours from 168 patients treated with chemotherapy with or without HER2 (encoded by ERBB2)-targeted therapy before surgery. Pathology end points (complete response or residual disease) at surgery3 were then correlated with multi-omic features in these diagnostic biopsies. Here we show that response to treatment is modulated by the pre-treated tumour ecosystem, and its multi-omics landscape can be integrated in predictive models using machine learning. The degree of residual disease following therapy is monotonically associated with pre-therapy features, including tumour mutational and copy number landscapes, tumour proliferation, immune infiltration and T cell dysfunction and exclusion. Combining these features into a multi-omic machine learning model predicted a pathological complete response in an external validation cohort (75 patients) with an area under the curve of 0.87. In conclusion, response to therapy is determined by the baseline characteristics of the totality of the tumour ecosystem captured through data integration and machine learning. This approach could be used to develop predictors for other cancers.

Residual cancer burden after neoadjuvant chemotherapy and long-term survival outcomes in breast cancer: a multicentre pooled analysis of 5161 patients.

BACKGROUND: Previous studies have independently validated the prognostic relevance of residual cancer burden (RCB) after neoadjuvant chemotherapy. We used results from several independent cohorts in a pooled patient-level analysis to evaluate the relationship of RCB with long-term prognosis across different phenotypic subtypes of breast cancer, to assess generalisability in a broad range of practice settings. METHODS: In this pooled analysis, 12 institutes and trials in Europe and the USA were identified by personal communications with site investigators. We obtained participant-level RCB results, and data on clinical and pathological stage, tumour subtype and grade, and treatment and follow-up in November, 2019, from patients (aged ≥18 years) with primary stage I-III breast cancer treated with neoadjuvant chemotherapy followed by surgery. We assessed the association between the continuous RCB score and the primary study outcome, event-free survival, using mixed-effects Cox models with the incorporation of random RCB and cohort effects to account for between-study heterogeneity, and stratification to account for differences in baseline hazard across cancer subtypes defined by hormone receptor status and HER2 status. The association was further evaluated within each breast cancer subtype in multivariable analyses incorporating random RCB and cohort effects and adjustments for age and pretreatment clinical T category, nodal status, and tumour grade. Kaplan-Meier estimates of event-free survival at 3, 5, and 10 years were computed for each RCB class within each subtype. FINDINGS: We analysed participant-level data from 5161 patients treated with neoadjuvant chemotherapy between Sept 12, 1994, and Feb 11, 2019. Median age was 49 years (IQR 20-80). 1164 event-free survival events occurred during follow-up (median follow-up 56 months [IQR 0-186]). RCB score was prognostic within each breast cancer subtype, with higher RCB score significantly associated with worse event-free survival. The univariable hazard ratio (HR) associated with one unit increase in RCB ranged from 1·55 (95% CI 1·41-1·71) for hormone receptor-positive, HER2-negative patients to 2·16 (1·79-2·61) for the hormone receptor-negative, HER2-positive group (with or without HER2-targeted therapy; p<0·0001 for all subtypes). RCB score remained prognostic for event-free survival in multivariable models adjusted for age, grade, T category, and nodal status at baseline: the adjusted HR ranged from 1·52 (1·36-1·69) in the hormone receptor-positive, HER2-negative group to 2·09 (1·73-2·53) in the hormone receptor-negative, HER2-positive group (p<0·0001 for all subtypes). INTERPRETATION: RCB score and class were independently prognostic in all subtypes of breast cancer, and generalisable to multiple practice settings. Although variability in hormone receptor subtype definitions and treatment across patients are likely to affect prognostic performance, the association we observed between RCB and a patient's residual risk suggests that prospective evaluation of RCB could be considered to become part of standard pathology reporting after neoadjuvant therapy. FUNDING: National Cancer Institute at the US National Institutes of Health.

Hyperpolarized Carbon-13 MRI for Early Response Assessment of Neoadjuvant Chemotherapy in Breast Cancer Patients.

Hyperpolarized 13C-MRI is an emerging tool for probing tissue metabolism by measuring 13C-label exchange between intravenously injected hyperpolarized [1-13C]pyruvate and endogenous tissue lactate. Here, we demonstrate that hyperpolarized 13C-MRI can be used to detect early response to neoadjuvant therapy in breast cancer. Seven patients underwent multiparametric 1H-MRI and hyperpolarized 13C-MRI before and 7-11 days after commencing treatment. An increase in the lactate-to-pyruvate ratio of approximately 20% identified three patients who, following 5-6 cycles of treatment, showed pathological complete response. This ratio correlated with gene expression of the pyruvate transporter MCT1 and lactate dehydrogenase A (LDHA), the enzyme catalyzing label exchange between pyruvate and lactate. Analysis of approximately 2,000 breast tumors showed that overexpression of LDHA and the hypoxia marker CAIX was associated with reduced relapse-free and overall survival. Hyperpolarized 13C-MRI represents a promising method for monitoring very early treatment response in breast cancer and has demonstrated prognostic potential. SIGNIFICANCE: Hyperpolarized carbon-13 MRI allows response assessment in patients with breast cancer after 7-11 days of neoadjuvant chemotherapy and outperformed state-of-the-art and research quantitative proton MRI techniques.