Using the Cancer Aging and Research Group- Breast Cancer (CARG-BC) predictive model in older adults (OA) with early breast cancer: an external validation study

Using the Cancer Aging and Research Group- Breast Cancer (CARG-BC) predictive model in older adults (OA) with early breast cancer: an external validation study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Using the Cancer Aging and Research Group- Breast Cancer (CARG-BC) predictive model in older adults (OA) with early breast cancer: an external validation study Neha Pathak, Ashley Kimmel, Yael Berner-Wygoda, Sulaiman A Almuthri, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7468335/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 02 Dec, 2025 Read the published version in Breast Cancer Research and Treatment → Version 1 posted 9 You are reading this latest preprint version Abstract Purpose: Decision-making for chemotherapy in early breast cancer (EBC) in OA (older adults: age ≥65 years) is complex due to frailty, multimorbidity, and competing risks for mortality. Magnuson (2021) developed a chemotherapy toxicity prediction score, CARG-BC; its external validation can improve generalizability. Objectives: CARG-BC’s ability to predict grade 3+ chemotoxicity in OA with EBC (primary), unplanned healthcare use, and changes to chemotherapy protocol (secondary). Methods: A single centre retrospective cohort study comprising OA with EBC who received (neo)adjuvant chemotherapy from 2013-2023. Clinical, demographic, CARG-BC, and healthcare usage variables were extracted from patient records. Risk groups based on CARG-BC score were compared using T-test (continuous variables) & χ2 test (categorical variables). Toxicity risk based on CARG-BC score was assessed using logistic regression. The predictive ability of the CARG-BC score was evaluated by calculating AUC. Results: Of 243 patients, the median age was 70 years (range 65-86), 99.6% female, 80.2% with comorbidities, 33.7% with polypharmacy, 28.8% living alone, and 8.2% seen in the geriatric oncology clinic. Over half (53.9%) had grade 3+ toxicities. Healthcare utilization included 19.8% of patients with at least one unplanned clinic visit, 29.6% an emergency care visit, and 14.4% a hospitalization. The median CARG-BC score was 7 (IQR 3, 8) and the CARG-BC AUC was 0.76 (95% Confidence interval [CI] 0.70, 0.82). The odds of grade 3+ toxicity is increased by 1.33 times per CARG-BC point increase. Conclusion: The CARG-BC model retained good discrimination for grade ≥3 chemotoxicity and should be used in shared-decision making with OA. breast cancer older adults geriatric oncology chemotherapy toxicity external validation Figures Figure 1 Figure 2 Figure 3 Introduction Breast cancer is the second most common cancer, the most common cancer in women, and accounts for 1 in 6 cancer deaths worldwide [1]. Breast cancer incidence increases with age, with over 40% of diagnoses occurring in adults who are 65 years and older, or older adults (OA) [2]. OA with breast cancer commonly have a prognostically favorable subtype of Estrogen/Progesterone receptor (ER/PR) positive and Human Epidermal Growth Factor-2 (HER2) negative cancer; however, up to 25-30% cases feature more aggressive subtypes of triple negative or HER2 positive disease, which often warrant chemotherapy [3]. The management of breast cancer in this population is nuanced, with the need to factor in life expectancy, comorbidities, functional status as well as patient preferences and values [4]. Further, OA have a higher incidence of chemotherapy related adverse effects (AE) compared to younger patients[5]. Clinical tools such as the PREDICT NHS [6] and molecular tests like the Oncotype Dx [7, 8] and MammaPrint [9] help estimate the benefits of chemotherapy in the curative setting. These have been tested in different patient populations and are widely used [10–12]. However, few tools exist for accurate estimation of the harms of chemotherapy, particularly severe or grade 3+ AE. Understanding the risk of severe AE is important as they can lead to functional decline, deterioration of quality of life, and unplanned healthcare use[5, 13, 14] In 2021, Magnuson et al., created a risk model in OA with early breast cancer named Cancer and Aging Research Group-Breast Cancer (CARG-BC) score, based on 8 clinical, laboratory and geriatric variables, which classified patients into low (22%), intermediate (51%) and high risk (81%) of grade ≥3 AE from chemotherapy (supplementary table 1) [15]. This model demonstrated good discriminatory ability, with an overall area under the receiver operating characteristic curve (AUC) of 0.73 and performed better than previous tumor-agnostic models (CARG score, AUC 0.56) and physicians’ clinical judgement. The score also correlated with treatment modifications, hospitalizations and a relative dose intensity (RDI) of <85%. While the authors developed and tested the CARG-BC model in separate development and validation cohorts, the patients in the latter were recruited from the same institutions [15]. External validation of prediction models can improve their generalizability, reproducibility and reliability, leading to wider uptake in clinical practice [16–18]. With this background, we designed this study to externally validate the CARG-BC score in a different patient population and healthcare setting. To our knowledge, there has been no report of external validation of this model. Materials and methods Study Objectives The primary objective was to validate the CARG-BC score’s ability to predict grade 3+ chemotoxicity in OA with early breast cancer. Secondary objectives included assessing the ability of the CARG-BC tool to predict chemotoxicity-related unplanned healthcare use (i.e., additional clinic, urgent care visits/emergency room visits and hospitalizations) and changes to chemotherapy protocol (upfront or downstream dose reductions, changes in recommendations from standard of care systemic therapy, and dose delays of >5 days and discontinuations). We also assessed the correlation of the CARG-BC score with RDI of <85% and its utility in prediction of grade 3+ toxicity in ‘younger old’ of 65-74 years group and ‘older old’, i.e. ≥75 years age group. We hypothesized that the CARG-BC tool could predict grade 3+ toxicity for older women with early breast cancer with moderate success (AUC 0.70, lower limit of the 95% confidence interval ≥0.6) in our patient population. Study design and Ethics We performed a retrospective chart review at the Princess Margaret Cancer Centre, Toronto, Canada. We received approval from the institute’s research ethics board prior to start of the study (Letter 23-5820). As this was a retrospective study, the need for informed patient consent was waived. Patient population and data extraction We included OA (adults 65 years and older) diagnosed with breast cancer staged I-III as per AJCC 7 th edition [19], between January 1, 2013 to October 31, 2023, and who received neoadjuvant/adjuvant curative intent chemotherapy. Patients were required to have received chemotherapy at our centre, to facilitate access to detailed electronic health records. Four authors (NP, AK, YB-W and SAA) extracted de-identified data, and all extracted data was reviewed by the primary author. The following data were extracted: demographic data (age, sex, comorbidity status as assessed by Charlson’s comorbidity index, weight, height, social predictors of health such as primary spoken language, working status (retired/full time/part time), education and living alone or with family) and clinical variables (breast cancer stage, receptor status, type of chemotherapy and regimen, number of cycles and dose per cycle, any grade ≥3 AE as per Common Terminology Criteria for Adverse Events (CTCAE) v 5.0 [20] from the start of chemotherapy and up until 3 months post the last cycle, as assessed by the research team, and baseline lab values of hemoglobin and abnormal liver function tests, as needed for the CARG-BC tool). We extracted the highest grade of a specific AE experienced by the patient throughout their chemotherapy. Patients could have multiple kinds of toxicity which were extracted. To estimate health care use, emergency room visits, urgent care and unplanned clinic visits, and hospitalizations were captured. Meetings were conducted every 2 weeks for the first three months and then as needed with the entire team to ensure concordance and resolve discrepancies. Initially, the chart review was done in pairs by extracting data for the same patients separately and then comparing. This was done for the first 30 patients and then randomly throughout remainder of data extraction. For chemotherapy protocol related modification definitions, see supplementary table 2. Patients did not routinely undergo a geriatric assessment. As such, the geriatric variables of the CARG-BC tool including any falls in the previous 6 months, anyone to give advice in a crisis and the ability to walk a mile were extrapolated from reviewed clinical notes and a record of the statements used as surrogate markers of these variables were maintained. For example, we estimated that a patient has someone to give advice in a crisis most of the time if they lived with a family member or family lived close by, they were noted to be ‘well supported with friends/family’, were accompanied to the important decision-making visits, and/or did not need social worker services. We also considered a priori that falls would have the most missingness and this was adjusted for in the analysis (see below). Statistical analysis Demographic and clinical characteristics, toxicity, RDI, and CARG-BC data were reported using descriptive statistics such as the mean, standard deviation (SD), median, interquartile range (Q1, Q3), range (min, max), frequency, or percentage, as appropriate. The CARG-BC score was summarized for all patients and compared between those with and without any grade 3+ toxicity using the Wilcoxon rank-sum test for scores and Fisher’ exact or Chi-squared tests for individual items. Logistic regression was used to evaluate the predictive ability of the CARG-BC score to identify the presence of CTCAE Grade 3 or higher toxicity using repeated cross-validation (5 folds, 5 repeats). Predictive performance was assessed through accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and the AUC. As a sensitivity analysis, the predictive performance of the CARG-BC score was reassessed after excluding the falls item. The relationship between RDI vs. CARG-BC score was assessed using correlation (Pearson) analysis and visualized through a scatterplot with a line-of-best fit based on linear regression. Patients who received the standard of care therapy were compared to those whose treatment deviated from the institutional standard using Wilcoxon rank-sum tests for continuous variables and Chi-squared or Fisher’s exact tests for categorical variables, to identify patients less likely to receive standard of care. A sample size of 243 patients was planned, which would have 80% power to detect an AUC of 0.70 at 0.05 level of significance. The assumed null AUC value for the ability of CARG-BC to predict grade 3+ chemotoxicity was 0.60. The assumed proportion of patients with grade 3+ chemotoxicity was estimated to be 46% based on the study by Magnuson et al [15]. The sample size was calculated using PASS 2023, version 23.0.2. Statistical analysis was performed using R Version 4.5.0 (R Core Team, 2024). Results Baseline characteristics Of the 243 patients included in our cohort, the median age was 70 years, with 82.3% in the (65-74) year age group. The majority were female (all but 1 of 243 patients), 70.8% were retired, 28.8% lived alone, and 33.7% took 5 or more prescription drugs at baseline. Stage distribution differed between treatment groups: only 1.0% of patients who received neoadjuvant chemotherapy had stage I disease, compared with 31.4% of those treated with adjuvant chemotherapy. We found 59.7% patients to have ER/PR+ status, 37.0% with HER2 driven disease (which could be ER/PR+ or ER/PR-) and 25.9% patients had triple negative breast cancer. The CARG-BC score was low in 93 (38.3%), intermediate in 134 (55.2%) and high in 16 (6.6%) patients, respectively (table 1). Docetaxel and Cyclophosphamide (TC) was the most commonly prescribed chemotherapy regimen 34.9% patients), followed by Trastuzumab with weekly Paclitaxel (i.e. the Tolaney regimen, 22.6%) and dose dense Adriamycin, Cyclophosphamide and Paclitaxel (dd AC-T, 22.2%). See figure 1 for details. Table 1 Baseline characteristics Characteristics Total n=243 Baseline characteristics Age, median (range), years 70 (65-86) ECOG PS ≥2 10 (4.1%) Comorbidities & 195 (80.2%) CCI, median (range) 0 (0,5) Body mass index, mean (SD) 27.6 (5.6) Body surface area, median (IQR) 1.7 (1.6, 1.8) Self-reported race/ethnicity White Black East Asian (China, Japan, Korea) South Asian Latin American Others Unavailable 44 (18.1%) 7 (2.9%) 20 (8.2%) 5 (2.1%) 1 (0.4%) 5 (2.1%) 161 (66.3%) Polypharmacy (≥5 drugs other than oncology drugs) 82 (33.7%) Seen in a geriatric oncology clinic 20 (8.2%) Social determinants of health Education Primary education Secondary education College Missing 5 (2.1%) 19 (7.8%) 84 (34.6%) 135 (55.6%) Living situation Alone With partner With family With roommates Missing 70 (28.8%) 122 (50.2%) 47 (19.3%) 1 (0.4%) 2 (1.2%) Working status Retired Full time Part time Never employed Missing 172 (70.8%) 32 (13.2%) 13 (5.3%) 5 (2.1%) 21 (8.6%) Clinical characteristics Stage (clinical, neoadjuvant) n=103 I 1 (1.0)% II 56 (54.4%) III 46 (44.7%) Stage (pathological, adjuvant) n=140 I 44 (31.4%) II 64 (45.7%) III 32 (22.9%) ER/PR+ and HER2 negative ER/PR+ and HER2 positive (triple positive) ER/PR negative and HER2 positive ER/PR negative and HER2 negative (triple negative) 90(37.0%) 55 (22.6%) 35 (14.4%) 63 (25.9%) CARG-BC score Low (0-5 points) Intermediate (6-11 points) High ( 12 points) 93 (38.3%) 134 (55.2%) 16 (6.6%) Abbreviations: CARG-BC, Cancer Aging and Research Group- Breast Cancer; CCI, Charlson’s comorbidity index; ECOG, Eastern Cooperative Oncology Group; ER, Estrogen receptor, HER2, human epidermal growth factor receptor 2; IQR, Inter Quartile Range; PR, Progesterone receptor, SD, standard deviation & Presence of any comorbidities Grade 3 and higher adverse effects Over half of the patients (53.9%) experienced grade ≥3 side effect. The most common non-hematological AEs were fatigue (20.2%), peripheral neuropathy (12.8%) and infection (7.8%), whereas the most common hematological AEs were neutropenia without fever (15.2%) and anemia (7.4%). Febrile neutropenia was seen in 4.9% of patients (figure 2, supplementary table 3). Primary objective Overall, the median CARG-BC score for this population was 7 (Inter Quartile Range [IQR], 3, 8). Each unit increase in CARG-BC score was associated with higher odds of grade ≥3 AE with an odds ratio (OR) of 1.33 (95% Confidence Interval [CI] =1.22, 1.46; p <0.001) and the AUC was 0.76 (95% CI = 0.70, 0.82). The observed risk of grade ≥3 AE in low, intermediate, and high-risk categories of CARG-BC was 28%, 67% and 94%, respectively (Figure 3a). The performance of CARG-BC in our cohort is comparable to the derivative cohort in the study by Magnuson et al [15], which had an OR of 1.28 (95% CI 1.19, 1.38, p<0.001) and AUC of 0.75 (95% CI 0.7, 0.81) in the development cohort and 0.69 (95% CI 0.62,0.77) in the validation cohort (Figure 3b). The details of each CARG-BC variable for the entire cohort is outlined in Table 2. The predictive ability of the CARG-BC score is displayed in Table 3. In the sensitivity analysis excluding the ‘falls’ item, the odds of grade 3+ toxicity and AUC were not different from the primary results (see supplementary table 4). Table 2 CARG-BC variables and their predictive performance for grade 3 adverse effects CARG-BC variable Full Sample (n=243) Grade ≥3 AE not experienced (n=112) Grade ≥3 AE experienced (n=131) p-value Anthracycline use 0.003 No 174 (71.6) 91 (81.2) 83 (63.4) Yes 69 (28.4) 21 (18.8) 48 (36.6) Stage II/III 0.006 No 46 (18.9) 30 (26.8) 16 (12.2) Yes 197 (81.1) 82 (73.2) 115 (87.8) Duration ≥12 weeks <0.001 No 136 (56.0) 81 (72.3) 55 (42.0) Yes 107 (44.0) 31 (27.7) 76 (58.0) Abnormal LFT 0.002 No 225 (92.6) 110 (98.2) 115 (87.8) Yes 18 (7.4) 2 (1.8) 16 (12.2) Hemoglobin less than cut off value $ 0.071 No 220 (90.5) 106 (94.6) 114 (87.0) Yes 23 (9.5) 6 (5.4) 17 (13.0) Falls 0.54 No 217 (89.3) 102 (91.1) 115 (87.8) Yes 26 (10.7) 10 (8.9) 16 (12.2) Ability to walk a mile 0.009 Limited 212 (87.2) 105 (93.8) 107 (81.7) Not limited 31 (12.8) 7 (6.2) 24 (18.3) Someone to give advice in a crisis 0.006 Most or all of the time 198 (81.5) 100 (89.3) 98 (74.8) None, little or some of the time 45 (18.5) 12 (10.7) 33 (25.2) Abbreviations: AE, adverse events; CARG-BC, Cancer and Aging Research Group-Breast Cancer; LFT, liver function tests $ Males ≤13 g/dl, Females ≤12 g/dl Table 3 Predictive ability of the CARG-BC score Predictive Performance of CARG-BC in Classifying Grade 3+ toxicity Measure Value (95% CI) Accuracy 70.8 (64.6, 76.4) Sensitivity 80.2 (72.3, 86.6) Specificity 59.8 (50.1, 69.0) PPV 70.0 (62.0, 77.2) NPV 72.0 (61.8, 80.9) AUC 0.759 (0.699, 0.818) OR 1.33 (1.22, 1.46); p <0.001 Abbreviations: AUC: Area Under the Curve of the Receiver Operating Characteristic; CI, confidence interval; NPV: Negative Predictive Value; OR, Odds Ratio; PPV: Positive Predictive Value Secondary objectives From the full cohort, 14.4% of patients were hospitalized, 29.6% visited the emergency or urgent care clinics, and 19.8% required an additional (non-routine) clinic appointment. There was a significant association between CARG-BC score and healthcare use outcomes including hospitalizations (OR 1.18 [95% CI 1.07, 1.31]; p 0.002) and urgent care use (OR 1.12 [95% CI 1.04, 1.22]; p 0.003), but did not meet statistical significance for unplanned clinic visits (supplementary table 5). The CARG-BC score was associated with increased odds of the composite outcome of healthcare utilization (all 3 variables combined) (OR 1.11 [95% CI 1.04, 1.20]; p 0.003). Healthcare utilization increased with increasing CARG-BC score grouping (supplementary figure 1). Chemotherapy protocol modifications occurred frequently. Dose delays and discontinuations were seen in 18.9% and 18.5% of patients, respectively and dose reductions occurred in 52.7%. Of these, 15.6% had upfront dose reductions, 69.5% began therapy at full dose and required subsequent dose reduction, and 14.8% had both upfront and subsequent dose reductions. The mean RDI was 87.0 (SD 18.9) and 30% patients had an RDI of <85%. Higher CARG-BC was associated with chemotherapy modifications including dose delay, dose reductions (whether upfront or downstream), and RDI <85% (supplementary table 6). There was a decrease of 0.61% in RDI per unit increase in the CARG-BC score but this was not statistically significant (supplementary figure 2). Patients who required a deviation from standard protocol as judged by the treating medical oncologist were compared to those who were able to initiate standard of care therapy. Older age, higher body mass index, more multimorbidity and non-working status of the patient were found to be associated with deviation from protocol, while CARG-BC score was not (supplementary table 7). In the 65–74-year age group, there was an OR of 1.37 (95%CI 1.24,1.53; p<0.001) and AUC Of 0.78 (95%CI 0.78, 0.84) for each increase in CARG-BC point. For patients aged 75 and older, slightly increased estimates of association and performance with OR of 1.55 [95%CI 1.16, 2.39]; p = 0.013) and AUC of 0.77 [95%CI 0.65, 0.9]). Discussion This study confirms the utility of the CARG-BC model in predicting grade 3 or higher toxicity in OA with early breast cancer, with reasonable accuracy. These results (OR 1.33 and AUC 0.76) are similar to the study by Magnuson et al (OR 1.28 in the developmental cohort and AUC of 0.73 in the overall cohort) [15]. Further, we were able to show an association of the CARG-BC score with hospitalizations, dose reductions, delays, and reduced RDI, similar to the original cohort [15]. To our knowledge, this study is the first external validation of this score, which can aid in clinical decision making for curative intent chemotherapy in this population. About 30% patients in our cohort visited the emergency room, and an increasing CARG-BC score was associated with greater emergency care use. Emergency room visits have a significant impact on the patient, family who often attend with them, and the healthcare system, so this information can add to the shared decision making between OA and their oncology team when selecting a treatment plan.. We saw fewer discontinuations than the study by Magnuson et al [15] (18.9% vs 24%), and the smaller numbers may explain the statistically non-significant association with the CARG-BC score in our cohort. The patient populations in the two cohorts were similar in terms of age, sex, living situation and working status. Our study had more stage II/III disease (81.1% vs 61%) which partly accounts for increased use of neoadjuvant chemotherapy (42.4% vs 17.3%). This may also reflect a shift medical oncology practice over time, with an increased use of neoadjuvant chemotherapy to assess biologic response and modify adjuvant therapy to improve cancer outcomes [21–24]. Our cohort comprised fewer patients with higher education and a higher proportion of non-White/Caucasian, which suggest that the predictive ability of CARG-BC score is accurate in diverse populations, although given the retrospective study (table 1) there was considerable missing data in several of these variables For the geriatric variable of falls,10.7% of patients in our cohort had a documented recent fall, similar to a 9% prevalence of falls in Magnuson et al [15]. As we anticipated missingness in this variable we incorporated a sensitivity analysis excluding this variable, but this did not reveal any difference. Nevertheless, the falls item was the least predictor of grade 3 or higher AEs among all variables in the CARG-BC score (table 2), likely because of poor quality documentation. In prospective studies that have validated the CARG score (which also includes a history of falls in the last six months as a variable), the prevalence of falls has been documented to be (7-18)% [25–28], with a lower incidence seen in patients with non-metastatic cancers. A recent systematic review estimated the prevalence of falls in OA with cancer at 24%, with a high degree of heterogeneity. Breast cancer was identified as a risk factor for falls in OA [29]. With this in mind, these results emphasize the value of incorporating falls in the history taking of an OA with breast cancer when planning for chemotherapy. While a statistically significant association between the CARG-BC score and healthcare utilization (urgent care and hospitalization) as well as chemotherapy modification (dose delay, dose reduction and reduced RDI) was seen in our study, the magnitude of these effects per unit increase of the CARG-BC score was small, unlike its correlation with the chance of grade 3 or higher AEs. As such, while there may be a clinically important difference between those who score low on the CARG-BC tool vs those who have a high score, small differences in scores between patients may not be as impactful. Conversely, calculating the CARG-BC score is straightforward, based on adding three geriatrics-related questions to the standard work up of a patient with early breast cancer. Other studies have indicated that the time taken to calculate a similar score such as the CARG score is <5 minutes and most oncologists have reported it to be worthwhile [30, 31]. The use of chemotherapy toxicity prediction scores such as the CARG-BC can eventually lead to less consumption of healthcare resources, distress for the patient and improve decision making to proceed with curative intent therapy, similar to the role played by the CARG score and a more extensive geriatric assessment [32–34]. Optimization of clinical and geriatric variables, such as choosing shorter or anthracycline free therapy is likely valuable in this context and meaningful to patients to reduce the potential harms of chemotherapy. Most studies that have validated the CARG score, which is used across various cancer subtypes and stages, have a prospective design, similar to the study by Hurria et al 28 . Magnuson et al developed the CARG-BC prospectively [15]. As our study was a retrospective study, we formulated a consensus on how to extrapolate the three geriatric variables in the CARG-BC tool from existing electronic health records. The successful validation of the CARG-BC score from reviewing health records may open up avenues of machine learning and automating the calculation of the CARG-BC score from clinical notes documented in the medical oncologists’ clinics, thus saving both time and resources [35–37]. When analyzing the CARG-BC score in age subgroups, we saw that it was similarly predictive of grade 3 or higher AEs in both 65-74 years and 75+ age group. Due to a much smaller cohort (n=43) and greater upfront chemotherapy modifications in the latter, however, the confidence interval of the OR was wide and crossed 1. Increasing age was seen to be a predictive factor for deviation from standard of care in our cohort. It is also possible that patients in this age group were only considered for chemotherapy if they were exceedingly fit to receive it, thus resulting in a selection bias. While the benefit of curative intent chemotherapy has been noted to be less in older patients with breast cancer, it is also true that fewer than 10% of patients older than 75 years are recruited to trials, thus limiting our understanding of the biology of breast cancer in this patient group [38, 39]. Ongoing studies are evaluating methods of de-escalation of chemotherapy for OA with early breast cancer, either by escalating the targeted therapy component (NCT04266249) or by replacing chemotherapy with targeted therapy, like cyclin dependent kinase 4/6 inhibitors (NCT03609047, NCT03644186). These could be a better tolerated alternative option for such patients in the future. This study had some limitations. This was a retrospective single centre study, and is susceptible to missing data and selection bias. We had fewer patients in the high risk CARG-BC group and in those 75 years and older. The highest grade of toxicity of each type across the entire duration of chemotherapy was considered, which may have led to those with longer chemotherapy duration with greater recorded toxicity. Data was extrapolated for the geriatric variables, which may have limited the quality of data extracted, however, we tried to mitigate this. To conclude, we have reinforced the value and validity of using the CARG-BC score in predicting severe chemotoxicity, emergency room visits, hospitalizations, reduced RDI and chemotherapy protocol modifications in OA with early breast cancer. The results of this study may be used to create machine learning models to extract this score from existing electronic health records, and its use will add to the shared decision making for curative intent chemotherapy in OA with early breast canc Declarations Funding sources: Based on this research study, the first author Neha Pathak received salary support as a clinical fellow through the Hold ’em for Life Oncology Fellowship. Conflict of interest: Michelle Nadler reports speaker honorarium and consulting fees from Novartis and Exact Sciences, outside of the scope of this submitted work. Neha Pathak received salary support as a clinical fellow through the Hold ’em for Life Oncology Fellowship, for purposes of conducting this study.The remaining authors have no conflicts with the present work. No authors report competing interests with present work Acknowledgements: Neha Pathak received salary support as a clinical fellow through the Hold ’em for Life Oncology Fellowship, for purposes of conducting this study. Authors contributions Neha Pathak : Conceptualization, Methodology, Data curation, Investigation, Formal analysis, Funding acquisition, Project administration, Writing – Original Draft Preparation . Ashley Kimmel: Data curation, Investigation, Formal analysis, Writing-Review and Editing . Yael Berner-Wygoda : Data curation, Investigation, Formal analysis, Writing-Review and Editing . Sulaiman A Almuthri : Data curation, Investigation, Formal analysis, Writing-Review and Editing . Anna Theresa Santiago : Methodology, Formal analysis, Funding acquisition, Visualization, Writing – Review and Editing . Rana Jin: Validation, Writing-Review and Editing. Susie Monginot: Validation, Writing-Review and Editing. Shabbir MH Alibhai: Conceptualization, Methodology, Data curation, Validation, Formal analysis, Supervision, Funding acquisition, Visualization, Resources, Writing – Review and Editing. Michelle B Nadler: Conceptualization, Methodology, Data curation, Validation, Formal analysis, Supervision, Funding acquisition, Visualization, Resources, Writing – Review and Editing Data availability statement: Data available on reasonable request to the corresponding author Ethics approval statement: This study received approval by the Institute’s Research Ethics Board (Letter 23-5820) prior to initiating the study Consent to participate: As this was a retrospective study, the need for informed patient consent was waived by the Institute’s Research Ethics Board. Consent to publish: As this was a retrospective study of deidentified data, the need for informed patient consent to publish was waived by the Institute’s Research Ethics Board. References Bray F, Laversanne M, Sung H, et al (2024) Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians 74:229–263. https://doi.org/10.3322/caac.21834 Cancer of the Breast (Female) - Cancer Stat Facts. In: SEER. https://seer.cancer.gov/statfacts/html/breast.html. Accessed 17 Feb 2024 Jenkins EO, Deal AM, Anders CK, et al (2014) Age-specific changes in intrinsic breast cancer subtypes: a focus on older women. Oncologist 19:1076–1083. https://doi.org/10.1634/theoncologist.2014-0184 Punglia RS, Hughes KS, Muss HB (2015) Management of Older Women with Early-Stage Breast Cancer. Am Soc Clin Oncol Educ Book 48–55. https://doi.org/10.14694/EdBook_AM.2015.35.48 Baltussen JC, de Glas NA, van Holstein Y, et al (2023) Chemotherapy-Related Toxic Effects and Quality of Life and Physical Functioning in Older Patients. JAMA Network Open 6:e2339116. https://doi.org/10.1001/jamanetworkopen.2023.39116 Grootes I, Wishart GC, Pharoah PDP (2024) An updated PREDICT breast cancer prognostic model including the benefits and harms of radiotherapy. npj Breast Cancer 10:6. https://doi.org/10.1038/s41523-024-00612-y Sparano JA, Gray RJ, Makower DF, et al (2018) Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. New England Journal of Medicine 379:111–121. https://doi.org/10.1056/NEJMoa1804710 Kalinsky K, Barlow WE, Gralow JR, et al (2021) 21-Gene Assay to Inform Chemotherapy Benefit in Node-Positive Breast Cancer. New England Journal of Medicine 385:2336–2347. https://doi.org/10.1056/NEJMoa2108873 Wittner BS, Sgroi DC, Ryan PD, et al (2008) Analysis of the MammaPrint Breast Cancer Assay in a Predominantly Postmenopausal Cohort. Clin Cancer Res 14:2988–2993. https://doi.org/10.1158/1078-0432.CCR-07-4723 Wishart GC, Bajdik CD, Azzato EM, et al (2011) A population-based validation of the prognostic model PREDICT for early breast cancer. European Journal of Surgical Oncology (EJSO) 37:411–417. https://doi.org/10.1016/j.ejso.2011.02.001 Hoskins KF, Danciu OC, Ko NY, Calip GS (2021) Association of Race/Ethnicity and the 21-Gene Recurrence Score With Breast Cancer–Specific Mortality Among US Women. JAMA Oncology 7:370–378. https://doi.org/10.1001/jamaoncol.2020.7320 Crager M, Wijayawardana SR, Gruver AM, et al (2022) Population-based estimate for the correlation of the Oncotype Dx Breast Recurrence Score® result and Ki-67 IHC MIB-1 pharmDx in HR+, HER2−, node-positive early breast cancer. Breast Cancer Res 24:74. https://doi.org/10.1186/s13058-022-01571-7 Battisti NML, Reed MWR, Herbert E, et al (2021) Bridging the Age Gap in breast cancer: Impact of chemotherapy on quality of life in older women with early breast cancer. European Journal of Cancer 144:269–280. https://doi.org/10.1016/j.ejca.2020.11.022 Walsh DJ, Sahm LJ, O’Driscoll M, et al (2023) Hospitalization due to adverse drug events in older adults with cancer: A retrospective analysis. J Geriatr Oncol 14:101540. https://doi.org/10.1016/j.jgo.2023.101540 Magnuson A, Sedrak MS, Gross CP, et al (2021) Development and Validation of a Risk Tool for Predicting Severe Toxicity in Older Adults Receiving Chemotherapy for Early-Stage Breast Cancer. J Clin Oncol 39:608–618. https://doi.org/10.1200/JCO.20.02063 Ramspek CL, Jager KJ, Dekker FW, et al (2020) External validation of prognostic models: what, why, how, when and where? Clin Kidney J 14:49–58. https://doi.org/10.1093/ckj/sfaa188 Siontis GCM, Tzoulaki I, Castaldi PJ, Ioannidis JPA (2015) External validation of new risk prediction models is infrequent and reveals worse prognostic discrimination. J Clin Epidemiol 68:25–34. https://doi.org/10.1016/j.jclinepi.2014.09.007 Debray TPA, Vergouwe Y, Koffijberg H, et al (2015) A new framework to enhance the interpretation of external validation studies of clinical prediction models. J Clin Epidemiol 68:279–289. https://doi.org/10.1016/j.jclinepi.2014.06.018 Edge SB, Compton CC (2010) The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol 17:1471–1474. https://doi.org/10.1245/s10434-010-0985-4 Common Terminology Criteria for Adverse Events (CTCAE). 155 Riedel F, Hoffmann AS, Moderow M, et al (2020) Time trends of neoadjuvant chemotherapy for early breast cancer. International Journal of Cancer 147:3049–3058. https://doi.org/10.1002/ijc.33122 Masuda N, Lee S-J, Ohtani S, et al (2017) Adjuvant Capecitabine for Breast Cancer after Preoperative Chemotherapy. New England Journal of Medicine 376:2147–2159. https://doi.org/10.1056/NEJMoa1612645 von Minckwitz G, Huang C-S, Mano MS, et al (2019) Trastuzumab Emtansine for Residual Invasive HER2-Positive Breast Cancer. New England Journal of Medicine 380:617–628. https://doi.org/10.1056/NEJMoa1814017 Baskin AS, Huppert LA, Kelil T, et al (2024) The neoadjuvant approach to treatment of breast cancer: Multidisciplinary management to improve outcomes. Surgical Oncology Insight 1:100059. https://doi.org/10.1016/j.soi.2024.100059 Uchiyama M, Miyazaki M, Hayashi T, et al (2024) Assessing the ability of the Cancer and Aging Research Group tool to predict chemotherapy toxicity in older Japanese patients: A prospective observational study. Journal of Geriatric Oncology 15:101814. https://doi.org/10.1016/j.jgo.2024.101814 Pang A, Jiali L, Ng A, et al (2022) Use of the Cancer and Aging Research Group Predictive Model for Chemotherapy-Related Toxic Effects in a Multiethnic, Older Adult Asian Population. JAMA Netw Open 5:e2237196. https://doi.org/10.1001/jamanetworkopen.2022.37196 Ostwal V, Ramaswamy A, Bhargava P, et al (2021) Cancer Aging Research Group (CARG) score in older adults undergoing curative intent chemotherapy: a prospective cohort study. BMJ Open 11:e047376. https://doi.org/10.1136/bmjopen-2020-047376 Hurria A, Togawa K, Mohile SG, et al (2011) Predicting Chemotherapy Toxicity in Older Adults With Cancer: A Prospective Multicenter Study. JCO 29:3457–3465. https://doi.org/10.1200/JCO.2011.34.7625 Lingamaiah D, Bushi G, Gaidhane S, et al (2025) Falls among geriatric cancer patients: a systematic review and meta-analysis of prevalence and risk across cancer types. BMC Geriatr 25:179. https://doi.org/10.1186/s12877-025-05722-1 Baxter MA, Rowe M, Zucker K, et al (2024) UK national observational cohort study investigating Tolerance of Anti-cancer Systemic Therapy in the Elderly: the TOASTIE study. bmjonc 3:. https://doi.org/10.1136/bmjonc-2024-000459 Mbewe A, Pike P, Lewis R, et al (2021) Implementing the cancer and aging research group (CARG) tool in the ambulatory oncology setting to drive informed treatment selection. JCO 39:209–209. https://doi.org/10.1200/JCO.2020.39.28_suppl.209 Hamaker ME, Wildes TM, Rostoft S (2017) Time to Stop Saying Geriatric Assessment Is Too Time Consuming. J Clin Oncol 35:2871–2874. https://doi.org/10.1200/JCO.2017.72.8170 Mintchev ME, Kalra AG, Kou C-TJ, et al Use of a Chemotherapy Toxicity Prediction Tool to Decrease Risks for Hospitalization in Older Patients. Cureus 14:e24465. https://doi.org/10.7759/cureus.24465 Mohile SG, Dale W, Somerfield MR, et al (2018) Practical Assessment and Management of Vulnerabilities in Older Patients Receiving Chemotherapy: ASCO Guideline for Geriatric Oncology. J Clin Oncol 36:2326–2347. https://doi.org/10.1200/JCO.2018.78.8687 Grim S, Kotz A, Kotz G, et al (2024) Development and validation of electronic health record-based, machine learning algorithms to predict quality of life among family practice patients. Sci Rep 14:30077. https://doi.org/10.1038/s41598-024-80064-3 Steele AJ, Denaxas SC, Shah AD, et al (2018) Machine learning models in electronic health records can outperform conventional survival models for predicting patient mortality in coronary artery disease. PLoS One 13:e0202344. https://doi.org/10.1371/journal.pone.0202344 Peterson DJ, Ostberg NP, Blayney DW, et al (2021) Machine Learning Applied to Electronic Health Records: Identification of Chemotherapy Patients at High Risk for Preventable Emergency Department Visits and Hospital Admissions. JCO Clin Cancer Inform 1106–1126. https://doi.org/10.1200/CCI.21.00116 Wang Z, Wang T, Xie Y, et al (2025) Unraveling the role of adjuvant chemotherapy in elderly triple-negative breast cancer: Insights from competing risk analysis using SEER data. Cancer Epidemiology 97:102853. https://doi.org/10.1016/j.canep.2025.102853 Hurria A, Dale W, Mooney M, et al (2014) Designing Therapeutic Clinical Trials for Older and Frail Adults With Cancer: U13 Conference Recommendations. J Clin Oncol 32:2587–2594. https://doi.org/10.1200/JCO.2013.55.0418 Additional Declarations Competing interest reported. Michelle Nadler reports speaker honorarium and consulting fees from Novartis and Exact Sciences, outside of the scope of this submitted work. Neha Pathak received salary support as a clinical fellow through the Hold ’em for Life Oncology Fellowship, for purposes of conducting this study.The remaining authors have no conflicts with the present work. No authors report competing interests with present work Supplementary Files SupplementaryCARGBCmanuscript.docx Cite Share Download PDF Status: Published Journal Publication published 02 Dec, 2025 Read the published version in Breast Cancer Research and Treatment → Version 1 posted Editorial decision: Revision requested 29 Oct, 2025 Reviews received at journal 29 Oct, 2025 Reviewers agreed at journal 13 Oct, 2025 Reviews received at journal 12 Oct, 2025 Reviewers agreed at journal 13 Sep, 2025 Reviewers invited by journal 11 Sep, 2025 Editor assigned by journal 27 Aug, 2025 Submission checks completed at journal 27 Aug, 2025 First submitted to journal 27 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7468335","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":515221239,"identity":"6c8096e2-eeef-4887-b6f4-fb6237a468c1","order_by":0,"name":"Neha Pathak","email":"","orcid":"","institution":"Princess Margaret Cancer Centre","correspondingAuthor":false,"prefix":"","firstName":"Neha","middleName":"","lastName":"Pathak","suffix":""},{"id":515221241,"identity":"7a1cefdf-4deb-41b0-9ee7-4452fbe2a56b","order_by":1,"name":"Ashley Kimmel","email":"","orcid":"","institution":"McMaster University","correspondingAuthor":false,"prefix":"","firstName":"Ashley","middleName":"","lastName":"Kimmel","suffix":""},{"id":515221246,"identity":"49f9b4ca-3d8a-4e81-9153-90a9fa695acf","order_by":2,"name":"Yael Berner-Wygoda","email":"","orcid":"","institution":"Princess Margaret Cancer Centre","correspondingAuthor":false,"prefix":"","firstName":"Yael","middleName":"","lastName":"Berner-Wygoda","suffix":""},{"id":515221248,"identity":"43d2d453-a89b-44ff-9796-62cade117527","order_by":3,"name":"Sulaiman A Almuthri","email":"","orcid":"","institution":"Princess Margaret Cancer Centre","correspondingAuthor":false,"prefix":"","firstName":"Sulaiman","middleName":"A","lastName":"Almuthri","suffix":""},{"id":515221249,"identity":"b5369ab2-de87-4b8d-9024-22eae9738bef","order_by":4,"name":"Anna Theresa Santiago","email":"","orcid":"","institution":"University Health Network","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"Theresa","lastName":"Santiago","suffix":""},{"id":515221250,"identity":"d707c2c5-629e-442f-91f9-faa0c2cc357f","order_by":5,"name":"Rana Jin","email":"","orcid":"","institution":"Princess Margaret Cancer Centre","correspondingAuthor":false,"prefix":"","firstName":"Rana","middleName":"","lastName":"Jin","suffix":""},{"id":515221252,"identity":"639c20fc-bc84-48bc-a24a-c587040fdfb7","order_by":6,"name":"Susie Monginot","email":"","orcid":"","institution":"Princess Margaret Cancer Centre","correspondingAuthor":false,"prefix":"","firstName":"Susie","middleName":"","lastName":"Monginot","suffix":""},{"id":515221254,"identity":"a4a46e0d-06a4-4df5-a4e2-bbc032b20cc5","order_by":7,"name":"Shabbir MH Alibhai","email":"","orcid":"","institution":"Princess Margaret Cancer Centre","correspondingAuthor":false,"prefix":"","firstName":"Shabbir","middleName":"MH","lastName":"Alibhai","suffix":""},{"id":515221255,"identity":"f625c590-004b-4ad6-9e8b-87f00dffa4c0","order_by":8,"name":"Michelle B Nadler","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+0lEQVRIie3PsWrCUBTG8SNCJtH1SEp8hRMCaQqFvsoJglkUBBcHwUyZtHP7FukbpAR0EV0DisRFFwcnySDiTUoLHbzp2OH+5/vj3A9ApfqHUQwaQN8BQAZMAQzAEvJUEBLPHjhXYJWSl+ibGF/E9cuIPa/u0xOhAUZ7l/Bw64XraQTZSEJi7dF8I7RA71gOLwa9cLPkymQmIzVNrxG6vs42ugH3wqRL1fy396KcXAjHfvNwRvfKHhXkWkLEfLH9KK74zAWpBDKi2c0JoRngceDwjM13QT6nrxKyiveYDZ9bDfQ+ktOIW/Wka6bZ+T756dfe6A9ApVKpVJJu/sJPO1sF+w4AAAAASUVORK5CYII=","orcid":"","institution":"Princess Margaret Cancer Centre","correspondingAuthor":true,"prefix":"","firstName":"Michelle","middleName":"B","lastName":"Nadler","suffix":""}],"badges":[],"createdAt":"2025-08-27 06:23:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7468335/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7468335/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10549-025-07869-2","type":"published","date":"2025-12-02T15:57:39+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":91843289,"identity":"28ca2c9c-2b32-4bf8-bafa-b5b2e59566d9","added_by":"auto","created_at":"2025-09-22 10:05:30","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9199037,"visible":true,"origin":"","legend":"","description":"","filename":"CARGBCmanuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/9f476f45be59d823799a2998.docx"},{"id":91843277,"identity":"a9fe9439-df65-4e0d-910c-439f5ed59e7b","added_by":"auto","created_at":"2025-09-22 10:05:30","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11013,"visible":true,"origin":"","legend":"","description":"","filename":"dd29f6aa52b24293a0d2befb5822aa1b.json","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/d58b7fa09c41ea6fcd066998.json"},{"id":91845632,"identity":"47741247-6da4-46f3-88ba-dd46cd29e8a5","added_by":"auto","created_at":"2025-09-22 10:13:30","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5984959,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryCARGBCmanuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/cc669028236b469d80f484dc.docx"},{"id":91843285,"identity":"cb1e62e2-ed64-4902-ae94-98cc590d0fb7","added_by":"auto","created_at":"2025-09-22 10:05:30","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":286876,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/a73fc7296cf631f1ad17bcbe.jpeg"},{"id":91845631,"identity":"21eafdc2-f4a7-4d95-963f-98393a3e7c61","added_by":"auto","created_at":"2025-09-22 10:13:30","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":233052,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/b1d90a1ab18d49f2321750d1.jpeg"},{"id":91843290,"identity":"f79e7797-dd86-417c-8bae-b8e5daed2ca6","added_by":"auto","created_at":"2025-09-22 10:05:30","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":183469,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/de1ee081b595c7ded917d15f.jpeg"},{"id":91843278,"identity":"de4ebc6e-fdf2-47a0-adbf-d62f2843ee4c","added_by":"auto","created_at":"2025-09-22 10:05:30","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1074,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/b9d639765d424565c9bad4f9.jpeg"},{"id":91845630,"identity":"7c4d60b8-7566-4a5d-aca8-3ad141080b26","added_by":"auto","created_at":"2025-09-22 10:13:30","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":66277,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/381f49398c4f2b70dfd25081.png"},{"id":91843281,"identity":"bd9aea9b-71d2-4332-b175-8a15289a9cfd","added_by":"auto","created_at":"2025-09-22 10:05:30","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":51873,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/e932ac0866c0c59197025c3c.png"},{"id":91843284,"identity":"c7f7de77-7d6e-4938-9b08-4c56091a3557","added_by":"auto","created_at":"2025-09-22 10:05:30","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41430,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/aa484a37c40298236a3725a6.png"},{"id":91847178,"identity":"22d3faca-8715-4f25-8883-979cb28188a8","added_by":"auto","created_at":"2025-09-22 10:21:30","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":935,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/7a746056b512dcf44915e171.png"},{"id":91843275,"identity":"cbbec34e-f4ad-46dc-a487-4d1b967fa046","added_by":"auto","created_at":"2025-09-22 10:05:30","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":116371,"visible":true,"origin":"","legend":"\u003cp\u003eChemotherapy regimens prescribed as neoadjuvant and adjuvant chemotherapy.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/e0b804c3ca55f103fc08235f.png"},{"id":91843288,"identity":"88d8f9dc-ae4c-4a65-a66f-8548f2c78c08","added_by":"auto","created_at":"2025-09-22 10:05:30","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":74123,"visible":true,"origin":"","legend":"\u003cp\u003eIncidence of the most common grade ≥3 adverse effects\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/e844067efe9f2bf9c2d90d8c.png"},{"id":91845629,"identity":"54cbdf60-898a-4999-b262-8959bda17fba","added_by":"auto","created_at":"2025-09-22 10:13:30","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":62112,"visible":true,"origin":"","legend":"\u003cp\u003eRisk of grade 3-5 adverse effects by CARG-BC score grouping into low (0-5), intermediate (6-11) and high (12+) groups in (a) this study and (b) in the validation cohort from the study by Magnuson et al., which derived the CARG-BC score [15].\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/4a48cd2acbffa4a54d4a0afb.png"},{"id":97724032,"identity":"6bc7ee93-c27f-45f1-8e78-869f73d62d92","added_by":"auto","created_at":"2025-12-08 16:11:13","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":930084,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/53a7497a-4b71-4186-9436-f50739731f54.pdf"},{"id":91843291,"identity":"54ae5549-21ee-48ef-a8e3-31b379833dcf","added_by":"auto","created_at":"2025-09-22 10:05:31","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":5984959,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryCARGBCmanuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7468335/v1/03355c3f3bd3cafc57549cf1.docx"}],"financialInterests":"Competing interest reported. Michelle Nadler reports speaker honorarium and consulting fees from Novartis and Exact Sciences, outside of the scope of this submitted work. Neha Pathak received salary support as a clinical fellow through the Hold ’em for Life Oncology Fellowship, for purposes of conducting this study.The remaining authors have no conflicts with the present work. No authors report competing interests with present work","formattedTitle":"Using the Cancer Aging and Research Group- Breast Cancer (CARG-BC) predictive model in older adults (OA) with early breast cancer: an external validation study","fulltext":[{"header":"Introduction","content":"\u003cp\u003eBreast cancer is the second most common cancer, the most common cancer in women, and accounts for 1 in 6 cancer deaths worldwide [1]. Breast cancer incidence increases with age, with over 40% of diagnoses occurring in adults who are 65 years and older, or older adults (OA) [2].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOA with breast cancer commonly have a prognostically favorable subtype of Estrogen/Progesterone receptor (ER/PR) positive and Human Epidermal Growth Factor-2 (HER2) negative cancer; however, up to 25-30% cases feature more aggressive subtypes of triple negative or HER2 positive disease, which often warrant chemotherapy [3]. The management of breast cancer in this population is nuanced, with the need to factor in life expectancy, comorbidities, functional status as well as patient preferences and values [4]. Further, OA have a higher incidence of chemotherapy related adverse effects (AE) compared to younger patients[5].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eClinical tools such as the PREDICT NHS [6] and molecular tests like the Oncotype Dx [7, 8] and MammaPrint [9] help estimate the benefits of chemotherapy in the curative setting. These have been tested in different patient populations and are widely used [10–12]. However, few tools exist for accurate estimation of the harms of chemotherapy, particularly severe or grade 3+ AE. Understanding the risk of severe AE is important as they can lead to functional decline, deterioration of quality of life, and unplanned healthcare use[5, 13, 14] In 2021, Magnuson et al., created a risk model in OA with early breast cancer named Cancer and Aging Research Group-Breast Cancer (CARG-BC) score, based on 8 clinical, laboratory and geriatric variables, which classified patients into low (22%), intermediate (51%) and high risk (81%) of grade ≥3 AE from chemotherapy (supplementary table 1) [15]. This model demonstrated good discriminatory ability, with an overall area under the receiver operating characteristic curve (AUC) of 0.73 and performed better than previous tumor-agnostic models (CARG score, AUC 0.56) and physicians’ clinical judgement. The score also correlated with treatment modifications, hospitalizations and a relative dose intensity (RDI) of \u0026lt;85%. While the authors developed and tested the CARG-BC model in separate development and validation cohorts, the patients in the latter were recruited from the same institutions [15]. External validation of prediction models can improve their generalizability, reproducibility and reliability, leading to wider uptake in clinical practice [16–18].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWith this background, we designed this study to externally validate the CARG-BC score in a different patient population and healthcare setting. To our knowledge, there has been no report of external validation of this model.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003e\u003cem\u003eStudy Objectives\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe primary objective was to validate the CARG-BC score’s ability to predict grade 3+ chemotoxicity in OA with early breast cancer.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSecondary objectives included assessing the ability of the CARG-BC tool to predict chemotoxicity-related unplanned healthcare use (i.e., additional clinic, urgent care visits/emergency room visits and hospitalizations) and changes to chemotherapy protocol (upfront or downstream dose reductions, changes in recommendations from standard of care systemic therapy, and dose delays of \u0026gt;5 days and discontinuations). We also assessed the correlation of the CARG-BC score with RDI of \u0026lt;85% and its utility in prediction of grade 3+ toxicity in ‘younger old’ of 65-74 years group and ‘older old’, i.e. ≥75 years age group.\u003c/p\u003e\n\u003cp\u003eWe hypothesized that the CARG-BC tool could predict grade 3+ toxicity for older women with early breast cancer with moderate success (AUC 0.70, lower limit of the 95% confidence interval ≥0.6) in our patient population.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eStudy design and Ethics\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe performed a retrospective chart review at the Princess Margaret Cancer Centre, Toronto, Canada. We received approval from the institute’s research ethics board prior to start of the study (Letter 23-5820). As this was a retrospective study, the need for informed patient consent was waived.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePatient population and data extraction\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe included OA (adults 65 years and older) diagnosed with breast cancer staged I-III as per AJCC 7\u003csup\u003eth\u003c/sup\u003e edition [19], between January 1, 2013 to October 31, 2023, and who received neoadjuvant/adjuvant curative intent chemotherapy. Patients were required to have received chemotherapy at our centre, to facilitate access to detailed electronic health records. Four authors (NP, AK, YB-W and SAA) extracted de-identified data, and all extracted data was reviewed by the primary author. The following data were extracted: demographic data (age, sex, comorbidity status as assessed by Charlson’s comorbidity index, weight, height, social predictors of health such as primary spoken language, working status (retired/full time/part time), education and living alone or with family) and clinical variables (breast cancer stage, receptor status, type of chemotherapy and regimen, number of cycles and dose per cycle, any grade ≥3 AE as per Common Terminology Criteria for Adverse Events (CTCAE) v 5.0 [20] from the start of chemotherapy and up until 3 months post the last cycle, as assessed by the research team, and baseline lab values of hemoglobin and abnormal liver function tests, as needed for the CARG-BC tool). We extracted the highest grade of a specific AE experienced by the patient throughout their chemotherapy. Patients could have multiple kinds of toxicity which were extracted. To estimate health care use, emergency room visits, urgent care and unplanned clinic visits, and hospitalizations were captured.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMeetings were conducted every 2 weeks for the first three months and then as needed with the entire team to ensure concordance and resolve discrepancies. Initially, the chart review was done in pairs by extracting data for the same patients separately and then comparing. This was done for the first 30 patients and then randomly throughout remainder of data extraction. For chemotherapy protocol related modification definitions, see supplementary table 2.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePatients did not routinely undergo a geriatric assessment. As such, the geriatric variables of the CARG-BC tool including any falls in the previous 6 months, anyone to give advice in a crisis and the ability to walk a mile were extrapolated from reviewed clinical notes and a record of the statements used as surrogate markers of these variables were maintained. For example, we estimated that a patient has someone to give advice in a crisis most of the time if they lived with a family member or family lived close by, they were noted to be ‘well supported with friends/family’, were accompanied to the important decision-making visits, and/or did not need social worker services. \u0026nbsp;We also considered a priori that falls would have the most missingness and this was adjusted for in the analysis (see below).\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003cem\u003eStatistical analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eDemographic and clinical characteristics, toxicity, RDI, and CARG-BC data were reported using descriptive statistics such as the mean, standard deviation (SD), median, interquartile range (Q1, Q3), range (min, max), frequency, or percentage, as appropriate. The CARG-BC score was summarized for all patients and compared between those with and without any grade 3+ toxicity using the Wilcoxon rank-sum test for scores and Fisher’ exact or Chi-squared tests for individual items.\u003c/p\u003e\n\u003cp\u003eLogistic regression was used to evaluate the predictive ability of the CARG-BC score to identify the presence of CTCAE Grade 3 or higher toxicity using repeated cross-validation (5 folds, 5 repeats). Predictive performance was assessed through accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and the AUC. As a sensitivity analysis, the predictive performance of the CARG-BC score was reassessed after excluding the falls item. The relationship between RDI vs. CARG-BC score was assessed using correlation (Pearson) analysis and visualized through a scatterplot with a line-of-best fit based on linear regression. \u0026nbsp; Patients who received the standard of care therapy were compared to those whose treatment deviated from the institutional standard using Wilcoxon rank-sum tests for continuous variables and Chi-squared or Fisher’s exact tests for categorical variables, to identify patients less likely to receive standard of care.\u003c/p\u003e\n\u003cp\u003eA sample size of 243 patients was planned, which would have 80% power to detect an AUC of 0.70 at 0.05 level of significance. The assumed null AUC value for the ability of CARG-BC to predict grade 3+ chemotoxicity was 0.60. The assumed proportion of patients with grade 3+ chemotoxicity was estimated to be 46% based on the study by Magnuson et al [15]. The sample size was calculated using PASS 2023, version 23.0.2. Statistical analysis was performed using R Version 4.5.0 (R Core Team, 2024).\u003cstrong\u003e\u003cbr\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cem\u003eBaseline characteristics\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOf the 243 patients included in our cohort, the median age was 70 years, with 82.3% in the (65-74) year age group. The majority were female (all but 1 of 243 patients), 70.8% were retired, 28.8% lived alone, and 33.7% took 5 or more prescription drugs at baseline. Stage distribution differed between treatment groups: only 1.0% of patients who received neoadjuvant chemotherapy had stage I disease, compared with 31.4% of those treated with adjuvant chemotherapy. We found 59.7% patients to have ER/PR+ status, 37.0% with HER2 driven disease (which could be ER/PR+ or ER/PR-) and 25.9% patients had triple negative breast cancer. The CARG-BC score was low in 93 (38.3%), intermediate in 134 (55.2%) and high in 16 (6.6%) patients, respectively (table 1). Docetaxel and Cyclophosphamide (TC) was the most commonly prescribed chemotherapy regimen 34.9% patients), followed by Trastuzumab with weekly Paclitaxel (i.e. the Tolaney regimen, 22.6%) and dose dense Adriamycin, Cyclophosphamide and Paclitaxel (dd AC-T, 22.2%). See figure 1 for details.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1\u003c/strong\u003e Baseline characteristics\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"574\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal n=243\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003e\u003cem\u003eBaseline characteristics\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eAge, median (range), years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e70 (65-86)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eECOG PS \u0026ge;2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e10 (4.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eComorbidities\u003csup\u003e\u0026amp;\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e195 (80.2%)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eCCI, median (range)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e0 (0,5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eBody mass index, mean (SD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e27.6 (5.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eBody surface area, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e1.7 (1.6, 1.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eSelf-reported race/ethnicity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 387px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; White\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Black\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; East Asian\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; (China, Japan, Korea)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; South Asian\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Latin American\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Others\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Unavailable\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; 44 (18.1%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;7 (2.9%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;20 (8.2%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;5 (2.1%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;1 (0.4%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;5 (2.1%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;161 (66.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003ePolypharmacy (\u0026ge;5 drugs other than oncology drugs)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e82 (33.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eSeen in a geriatric oncology clinic\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e20 (8.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003e\u003cem\u003eSocial determinants of health\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eEducation\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" style=\"width: 73px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 340px;\"\u003e\n \u003cp\u003ePrimary education\u003c/p\u003e\n \u003cp\u003eSecondary education\u003c/p\u003e\n \u003cp\u003eCollege\u003c/p\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003e5 (2.1%)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e19 (7.8%)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e84 (34.6%)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e135 (55.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eLiving situation\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" style=\"width: 73px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 340px;\"\u003e\n \u003cp\u003eAlone\u003c/p\u003e\n \u003cp\u003eWith partner\u003c/p\u003e\n \u003cp\u003eWith family\u003c/p\u003e\n \u003cp\u003eWith roommates\u003c/p\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003e70 (28.8%)\u003c/p\u003e\n \u003cp\u003e122 (50.2%)\u003c/p\u003e\n \u003cp\u003e47 (19.3%)\u003c/p\u003e\n \u003cp\u003e1 (0.4%)\u003c/p\u003e\n \u003cp\u003e2 (1.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eWorking status\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" style=\"width: 73px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 340px;\"\u003e\n \u003cp\u003eRetired\u003c/p\u003e\n \u003cp\u003eFull time\u003c/p\u003e\n \u003cp\u003ePart time\u003c/p\u003e\n \u003cp\u003eNever employed\u003c/p\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003e172 (70.8%)\u003c/p\u003e\n \u003cp\u003e32 (13.2%)\u003c/p\u003e\n \u003cp\u003e13 (5.3%)\u003c/p\u003e\n \u003cp\u003e5 (2.1%)\u003c/p\u003e\n \u003cp\u003e21 (8.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003e\u003cem\u003eClinical characteristics\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eStage (clinical, neoadjuvant)\u003c/p\u003e\n \u003cp\u003en=103\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003eI \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; 1 (1.0)%\u003c/p\u003e\n \u003cp\u003eII \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;56 (54.4%)\u003c/p\u003e\n \u003cp\u003eIII \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; 46 (44.7%)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eStage (pathological, adjuvant)\u003c/p\u003e\n \u003cp\u003en=140\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003eI \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; 44 (31.4%)\u003c/p\u003e\n \u003cp\u003eII \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;64 (45.7%)\u003c/p\u003e\n \u003cp\u003eIII \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; 32 (22.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eER/PR+ and HER2 negative\u003c/p\u003e\n \u003cp\u003eER/PR+ and HER2 positive (triple positive)\u003c/p\u003e\n \u003cp\u003eER/PR negative and HER2 positive\u003c/p\u003e\n \u003cp\u003eER/PR negative and HER2 negative (triple negative)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e90(37.0%)\u003c/p\u003e\n \u003cp\u003e55 (22.6%)\u003c/p\u003e\n \u003cp\u003e35 (14.4%)\u003c/p\u003e\n \u003cp\u003e63 (25.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 382px;\"\u003e\n \u003cp\u003eCARG-BC score\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Low (0-5 points)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Intermediate (6-11 points)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; High (\u003cimg width=\"11\" height=\"17\" src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAaCAMAAACJtiw1AAAAAXNSR0IArs4c6QAAAE5QTFRFAAAAAAAAAAA6ADo6ADpmADqQOgAAOjpmOmaQOma2ZjoAZpC2ZrbbkGY6kLbbtpBmttvb25Bm27aQ29u229v/2////7Zm/7aQ/9vb///bsIoMPAAAAAF0Uk5TAEDm2GYAAAAJcEhZcwAAFiUAABYlAUlSJPAAAAAZdEVYdFNvZnR3YXJlAE1pY3Jvc29mdCBPZmZpY2V/7TVxAAAATUlEQVQoU2NgGCxAnFsU1SmS/GzsAmiuE+Zl4hRBUybExcyDplWCn40DXasgGyuSKlQVkkJcLHwIWVRbMNwhjm4dRhCKMSIBZGupGNgAgfQDJBSPoZkAAAAASUVORK5CYII=\" alt=\"image\"\u003e12 points)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 166px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e93 (38.3%)\u003c/p\u003e\n \u003cp\u003e134 (55.2%)\u003c/p\u003e\n \u003cp\u003e16 (6.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 26px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"6\" valign=\"top\" style=\"width: 548px;\"\u003e\n \u003cp\u003eAbbreviations: CARG-BC,\u0026nbsp;Cancer Aging and Research Group- Breast Cancer; CCI, Charlson\u0026rsquo;s comorbidity index; ECOG, Eastern Cooperative Oncology Group; ER, Estrogen receptor, HER2, human epidermal growth factor receptor 2; IQR, Inter Quartile Range; PR, Progesterone receptor, SD, standard deviation\u003c/p\u003e\n \u003cp\u003e\u003csup\u003e\u0026amp;\u003c/sup\u003e Presence of any comorbidities\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003eGrade 3 and higher adverse effects\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOver half of the patients (53.9%) experienced grade \u0026ge;3 side effect. The most common non-hematological AEs were fatigue (20.2%), peripheral neuropathy (12.8%) and infection (7.8%), whereas the most common hematological AEs were neutropenia without fever (15.2%) and anemia (7.4%). Febrile neutropenia was seen in 4.9% of patients (figure 2, supplementary table 3).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePrimary objective\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOverall, the median CARG-BC score for this population was 7 (Inter Quartile Range [IQR], 3, 8). Each unit increase in CARG-BC score was associated with higher odds of grade \u0026ge;3 AE with an odds ratio (OR) of 1.33 (95% Confidence Interval [CI] =1.22, 1.46; p \u0026lt;0.001) and the AUC was 0.76 (95% CI = 0.70, 0.82). The observed risk of grade \u0026ge;3 AE in low, intermediate, and high-risk categories of CARG-BC was 28%, 67% and 94%, respectively (Figure 3a). The performance of CARG-BC in our cohort is comparable to the derivative cohort in the study by Magnuson et al [15], which had an OR \u0026nbsp;of 1.28 (95% CI 1.19, 1.38, p\u0026lt;0.001) and AUC of 0.75 (95% CI 0.7, 0.81) in the development cohort and 0.69 (95% CI 0.62,0.77) in the validation cohort \u0026nbsp;(Figure 3b).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe details of each CARG-BC variable for the entire cohort is outlined in Table 2. The predictive ability of the CARG-BC score is displayed in Table 3. In the sensitivity analysis excluding the \u0026lsquo;falls\u0026rsquo; item, the odds of grade 3+ toxicity and AUC \u0026nbsp;were not different from the primary results (see supplementary table 4).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e CARG-BC variables and their predictive performance for grade\u003cimg width=\"11\" height=\"17\" src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAaCAMAAACJtiw1AAAAAXNSR0IArs4c6QAAAE5QTFRFAAAAAAAAAAA6ADo6ADpmADqQOgAAOjpmOmaQOma2ZjoAZpC2ZrbbkGY6kLbbtpBmttvb25Bm27aQ29u229v/2////7Zm/7aQ/9vb///bsIoMPAAAAAF0Uk5TAEDm2GYAAAAJcEhZcwAAFiUAABYlAUlSJPAAAAAZdEVYdFNvZnR3YXJlAE1pY3Jvc29mdCBPZmZpY2V/7TVxAAAATUlEQVQoU2NgGCxAnFsU1SmS/GzsAmiuE+Zl4hRBUybExcyDplWCn40DXasgGyuSKlQVkkJcLHwIWVRbMNwhjm4dRhCKMSIBZGupGNgAgfQDJBSPoZkAAAAASUVORK5CYII=\" alt=\"image\"\u003e3 adverse effects\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;CARG-BC variable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFull Sample (n=243)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGrade \u0026ge;3 AE not experienced (n=112)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGrade \u0026ge;3 AE experienced (n=131)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003eAnthracycline use\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e174 (71.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e91 (81.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e83 (63.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e69 (28.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e21 (18.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e48 (36.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003eStage II/III\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.006\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;No\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e46 (18.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e30 (26.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e16 (12.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e197 (81.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e82 (73.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e115 (87.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003eDuration \u0026ge;12 weeks\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e136 (56.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e81 (72.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e55 (42.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e107 (44.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e31 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e76 (58.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003eAbnormal LFT\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e225 (92.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e110 (98.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e115 (87.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e18 (7.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e2 (1.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e16 (12.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003eHemoglobin less than cut off value\u003csup\u003e$\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.071\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e220 (90.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e106 (94.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e114 (87.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e23 (9.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e6 (5.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e17 (13.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003eFalls\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e217 (89.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e102 (91.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e115 (87.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e26 (10.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e10 (8.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e16 (12.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003eAbility to walk a mile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.009\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Limited\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e212 (87.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e105 (93.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e107 (81.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Not limited\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e31 (12.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e7 (6.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e24 (18.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003eSomeone to give advice in a crisis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e0.006\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Most or all of the time\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e198 (81.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e100 (89.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e98 (74.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;None, little or some of the time\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 142px;\"\u003e\n \u003cp\u003e45 (18.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e12 (10.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e33 (25.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 82px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 602px;\"\u003e\n \u003cp\u003eAbbreviations: AE, adverse events; CARG-BC, Cancer and Aging Research Group-Breast Cancer; LFT, liver function tests\u003c/p\u003e\n \u003cp\u003e\u003csup\u003e$\u003c/sup\u003e Males \u0026le;13 g/dl, Females \u0026le;12 g/dl\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3\u003c/strong\u003e Predictive ability of the CARG-BC score\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 491px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePredictive Performance of CARG-BC in Classifying Grade 3+ toxicity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 192px;\"\u003e\n \u003cp\u003eMeasure\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 299px;\"\u003e\n \u003cp\u003eValue (95% CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 192px;\"\u003e\n \u003cp\u003eAccuracy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 299px;\"\u003e\n \u003cp\u003e70.8 (64.6, 76.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 192px;\"\u003e\n \u003cp\u003eSensitivity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 299px;\"\u003e\n \u003cp\u003e80.2 (72.3, 86.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 192px;\"\u003e\n \u003cp\u003eSpecificity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 299px;\"\u003e\n \u003cp\u003e59.8 (50.1, 69.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 192px;\"\u003e\n \u003cp\u003ePPV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 299px;\"\u003e\n \u003cp\u003e70.0 (62.0, 77.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 192px;\"\u003e\n \u003cp\u003eNPV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 299px;\"\u003e\n \u003cp\u003e72.0 (61.8, 80.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 192px;\"\u003e\n \u003cp\u003eAUC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 299px;\"\u003e\n \u003cp\u003e0.759 (0.699, 0.818)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 192px;\"\u003e\n \u003cp\u003eOR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 299px;\"\u003e\n \u003cp\u003e1.33 (1.22, 1.46); p \u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 491px;\"\u003e\n \u003cp\u003eAbbreviations: AUC: Area Under the Curve of the Receiver Operating Characteristic; CI, confidence interval; NPV: Negative Predictive Value; OR, Odds Ratio; PPV: Positive Predictive Value \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003eSecondary objectives\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eFrom the full cohort, 14.4% of patients were hospitalized, 29.6% visited the emergency or urgent care clinics, and 19.8% required an additional (non-routine) clinic appointment. There was a significant association between CARG-BC score and healthcare use outcomes including hospitalizations (OR 1.18 [95% CI 1.07, 1.31]; p 0.002) and urgent care use (OR 1.12 [95% CI 1.04, 1.22]; p 0.003), but did not meet statistical significance for unplanned clinic visits (supplementary table 5). The CARG-BC score was associated with increased odds of the composite outcome of healthcare utilization (all 3 variables combined) (OR 1.11 [95% CI 1.04, 1.20]; p 0.003). Healthcare utilization increased with increasing CARG-BC score grouping (supplementary figure 1).\u003c/p\u003e\n\u003cp\u003eChemotherapy protocol modifications occurred frequently. Dose delays and discontinuations were seen in 18.9% and 18.5% of patients, respectively and dose reductions occurred in 52.7%. Of these, 15.6% had upfront dose reductions, 69.5% began therapy at full dose and required subsequent dose reduction, and 14.8% had both upfront and subsequent dose reductions. The mean RDI was 87.0 (SD 18.9) and 30% patients had an RDI of \u0026lt;85%. \u0026nbsp;Higher CARG-BC was associated with chemotherapy modifications including dose delay, dose reductions (whether upfront or downstream), and RDI \u0026lt;85% (supplementary table 6). There was a decrease of 0.61% in RDI per unit increase in the CARG-BC score but this was not statistically significant (supplementary figure 2). Patients who required a deviation from standard protocol as judged by the treating medical oncologist were compared to those who were able to initiate standard of care therapy. Older age, higher body mass index, more multimorbidity and non-working status of the patient were found to be associated with deviation from protocol, while CARG-BC score was not (supplementary table 7).\u003c/p\u003e\n\u003cp\u003eIn the 65\u0026ndash;74-year age group, there was an OR of 1.37 (95%CI 1.24,1.53; p\u0026lt;0.001) and AUC Of 0.78 (95%CI 0.78, 0.84) for each increase in CARG-BC point. For patients aged 75 and older, slightly increased estimates of association and performance with OR of 1.55 [95%CI 1.16, 2.39]; p = 0.013) and AUC of 0.77 [95%CI 0.65, 0.9]).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study confirms the utility of the CARG-BC model in predicting grade 3 or higher toxicity in OA with early breast cancer, with reasonable accuracy. These results (OR 1.33 and AUC 0.76) \u0026nbsp; are similar to the study by Magnuson et al (OR 1.28 in the developmental cohort and AUC of 0.73 in the overall cohort) [15]. Further, we were able to show an association of the CARG-BC score with hospitalizations, dose reductions, delays, and reduced RDI, similar to the original cohort [15]. To our knowledge, this study is the first external validation of this score, which can aid in clinical decision making for curative intent chemotherapy in this population.\u003c/p\u003e\n\u003cp\u003eAbout 30% patients in our cohort visited the emergency room, and an increasing CARG-BC score was associated with greater emergency care use. Emergency room visits have a significant impact on the patient, family who often attend with them, and the healthcare system, so this information can add to the shared decision making between OA and their oncology team when selecting a treatment plan.. We saw fewer discontinuations than the study by Magnuson et al [15] (18.9% vs 24%), and the smaller numbers may explain the statistically non-significant association with the CARG-BC score in our cohort.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe patient populations in the two cohorts were similar in terms of age, sex, living situation and working status. Our study had more stage II/III disease (81.1% vs 61%) which partly accounts for increased use of neoadjuvant chemotherapy (42.4% vs 17.3%). This may also reflect a shift medical oncology practice over time, with an increased use of neoadjuvant chemotherapy to assess biologic response and modify adjuvant therapy to improve cancer outcomes [21\u0026ndash;24]. Our cohort comprised fewer patients with higher education and a higher proportion of non-White/Caucasian, which suggest that the predictive ability of CARG-BC score is accurate in diverse populations, although given the retrospective study (table 1) there was considerable missing data in several of these variables For the geriatric variable of falls,10.7% of patients in our cohort had a documented recent fall, similar to a 9% prevalence of falls in Magnuson et al [15]. As we anticipated missingness in this variable we incorporated a sensitivity analysis excluding this variable, but this did not reveal any difference. \u0026nbsp;Nevertheless, the falls item was the least predictor of grade 3 or higher AEs among all variables in the CARG-BC score (table 2), likely because of poor quality documentation. \u0026nbsp;In prospective studies that have validated the CARG score (which also includes a history of falls in the last six months as a variable), the prevalence of falls has been documented to be \u0026nbsp;(7-18)% [25\u0026ndash;28], with a lower incidence seen \u0026nbsp;in patients with non-metastatic cancers. \u0026nbsp;A recent systematic review estimated the prevalence of falls in OA with cancer at 24%, with a high degree of heterogeneity. Breast cancer was identified as a risk factor for falls in OA [29]. With this in mind, these results emphasize the value of incorporating falls in the history taking of an OA with breast cancer when planning for chemotherapy.\u003c/p\u003e\n\u003cp\u003eWhile a statistically significant association between the CARG-BC score and healthcare utilization (urgent care and hospitalization) as well as chemotherapy modification (dose delay, dose reduction and reduced RDI) was seen in our study, the magnitude of these effects per unit increase of the CARG-BC score was small, unlike its correlation with the chance of grade 3 or higher AEs. As such, while there may be a clinically important difference between those who score low on the CARG-BC tool vs those who have a high score, small differences in scores between patients may not be as impactful. Conversely, calculating the CARG-BC score is straightforward, based on adding three geriatrics-related questions to the standard work up of a patient with early breast cancer. Other studies have indicated that the time taken to calculate a similar score such as the CARG score is \u0026lt;5 minutes and most oncologists have reported it to be worthwhile [30, 31]. \u0026nbsp;The use of chemotherapy toxicity prediction scores such as the CARG-BC can eventually lead to less consumption of healthcare resources, distress for the patient and improve decision making to proceed with curative intent therapy, similar to the role played by the CARG score and a more extensive geriatric assessment [32\u0026ndash;34]. Optimization of clinical and geriatric variables, such as choosing shorter or anthracycline free therapy is likely valuable in this context and meaningful to patients to reduce the potential harms of chemotherapy.\u003c/p\u003e\n\u003cp\u003eMost studies that have validated the CARG score, which is used across various cancer subtypes and stages, have a prospective design, similar to the study by Hurria et al\u0026nbsp;\u003csup\u003e28\u003c/sup\u003e. Magnuson et al developed the CARG-BC prospectively [15]. \u0026nbsp;As our study was a retrospective study, we formulated a consensus on how to extrapolate the three geriatric variables in the CARG-BC tool from existing electronic health records. The successful validation of the CARG-BC score from reviewing health records may open up avenues of machine learning and automating the calculation of the CARG-BC score from clinical notes documented in the medical oncologists\u0026rsquo; clinics, thus saving both time and resources [35\u0026ndash;37].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhen analyzing the CARG-BC score in age subgroups, we saw that it was similarly predictive of grade 3 or higher AEs in both 65-74 years and 75+ \u0026nbsp;age group. Due to a much smaller cohort (n=43) and \u0026nbsp;greater upfront chemotherapy modifications in the latter, however, the confidence interval of the OR was wide and crossed 1. Increasing age was seen to be a predictive factor for deviation from standard of care in our cohort. It is also possible that patients in this age group were only considered for chemotherapy if they were exceedingly fit to receive it, thus resulting in a selection bias. While the benefit of curative intent chemotherapy has been noted to be less in older patients with breast cancer, it is also true that fewer than 10% of patients older than 75 years are recruited to trials, thus limiting our understanding of the biology of breast cancer in this patient group [38, 39]. Ongoing studies are evaluating methods of de-escalation of chemotherapy for OA with early breast cancer, either by escalating the targeted therapy component (NCT04266249) or by replacing chemotherapy with targeted therapy, like cyclin dependent kinase 4/6 inhibitors (NCT03609047, NCT03644186).\u0026nbsp;These could be a better tolerated alternative option for such patients in the future.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study had some limitations. This was a retrospective single centre study, and is susceptible to missing data and selection bias. We had fewer patients in the high risk CARG-BC group and in those 75 years and older. The highest grade of toxicity of each type across the entire duration of chemotherapy was considered, which may have led to those with longer chemotherapy duration with greater recorded toxicity. Data was extrapolated for the geriatric variables, which may have limited the quality of data extracted, however, we tried to mitigate this.\u003c/p\u003e\n\u003cp\u003eTo conclude, we have reinforced the value and validity of using the CARG-BC score in predicting severe chemotoxicity, emergency room visits, hospitalizations, reduced RDI and chemotherapy protocol modifications in OA with early breast cancer. The results of this study may be used to create machine learning models to extract this score from existing electronic health records, and its use will add to the shared decision making for curative intent chemotherapy in OA with early breast canc\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cem\u003eFunding sources:\u003c/em\u003eBased on this research study, the first author Neha Pathak received salary support as a clinical fellow through the Hold ’em for Life Oncology Fellowship.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eConflict of interest:\u003c/em\u003eMichelle Nadler reports speaker honorarium and consulting fees from Novartis and Exact Sciences, outside of the scope of this submitted work. Neha Pathak received salary support as a clinical fellow through the Hold ’em for Life Oncology Fellowship, for purposes of conducting this study.The remaining authors have no conflicts with the present work. No authors report competing interests with present work\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAcknowledgements:\u003c/em\u003eNeha Pathak received salary support as a clinical fellow through the Hold ’em for Life Oncology Fellowship, for purposes of conducting this study.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAuthors contributions\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNeha Pathak\u003c/strong\u003e: Conceptualization, Methodology, Data curation, Investigation, Formal analysis, Funding acquisition, Project administration, Writing – Original Draft Preparation\u003cstrong\u003e. Ashley Kimmel:\u003c/strong\u003e Data curation, Investigation, Formal analysis, Writing-Review and Editing\u003cstrong\u003e. Yael Berner-Wygoda\u003c/strong\u003e: Data curation, Investigation, Formal analysis, Writing-Review and Editing\u003cstrong\u003e. Sulaiman A Almuthri\u003c/strong\u003e: Data curation, Investigation, Formal analysis, Writing-Review and Editing\u003cstrong\u003e. Anna Theresa Santiago\u003c/strong\u003e: Methodology, Formal analysis, Funding acquisition, Visualization, Writing – Review and Editing\u003cstrong\u003e. Rana Jin:\u0026nbsp;\u003c/strong\u003eValidation, Writing-Review and Editing. \u003cstrong\u003eSusie Monginot:\u003c/strong\u003e Validation, Writing-Review and Editing.\u003cstrong\u003eShabbir MH Alibhai:\u003c/strong\u003e Conceptualization, Methodology, Data curation, Validation, Formal analysis, Supervision, Funding acquisition, Visualization, Resources, Writing – Review and Editing. \u003cstrong\u003eMichelle B Nadler:\u003c/strong\u003e Conceptualization, Methodology, Data curation, Validation, Formal analysis, Supervision, Funding acquisition, Visualization, Resources, Writing – Review and Editing\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eData availability statement:\u003c/em\u003eData available on reasonable request to the corresponding author\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEthics approval statement:\u003c/em\u003eThis study received approval by the Institute’s Research Ethics Board (Letter 23-5820) prior to initiating the study\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eConsent to participate:\u003c/em\u003eAs this was a retrospective study, the need for informed patient consent was waived by the Institute’s Research Ethics Board.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eConsent to publish:\u003c/em\u003e As this was a retrospective study of deidentified data, the need for informed patient consent to publish was waived by the Institute’s Research Ethics Board.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBray F, Laversanne M, Sung H, et al (2024) Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians 74:229\u0026ndash;263. https://doi.org/10.3322/caac.21834\u003c/li\u003e\n\u003cli\u003eCancer of the Breast (Female) - Cancer Stat Facts. In: SEER. https://seer.cancer.gov/statfacts/html/breast.html. Accessed 17 Feb 2024\u003c/li\u003e\n\u003cli\u003eJenkins EO, Deal AM, Anders CK, et al (2014) Age-specific changes in intrinsic breast cancer subtypes: a focus on older women. Oncologist 19:1076\u0026ndash;1083. https://doi.org/10.1634/theoncologist.2014-0184\u003c/li\u003e\n\u003cli\u003ePunglia RS, Hughes KS, Muss HB (2015) Management of Older Women with Early-Stage Breast Cancer. Am Soc Clin Oncol Educ Book 48\u0026ndash;55. https://doi.org/10.14694/EdBook_AM.2015.35.48\u003c/li\u003e\n\u003cli\u003eBaltussen JC, de Glas NA, van Holstein Y, et al (2023) Chemotherapy-Related Toxic Effects and Quality of Life and Physical Functioning in Older Patients. JAMA Network Open 6:e2339116. https://doi.org/10.1001/jamanetworkopen.2023.39116\u003c/li\u003e\n\u003cli\u003eGrootes I, Wishart GC, Pharoah PDP (2024) An updated PREDICT breast cancer prognostic model including the benefits and harms of radiotherapy. npj Breast Cancer 10:6. https://doi.org/10.1038/s41523-024-00612-y\u003c/li\u003e\n\u003cli\u003eSparano JA, Gray RJ, Makower DF, et al (2018) Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. New England Journal of Medicine 379:111\u0026ndash;121. https://doi.org/10.1056/NEJMoa1804710\u003c/li\u003e\n\u003cli\u003eKalinsky K, Barlow WE, Gralow JR, et al (2021) 21-Gene Assay to Inform Chemotherapy Benefit in Node-Positive Breast Cancer. New England Journal of Medicine 385:2336\u0026ndash;2347. https://doi.org/10.1056/NEJMoa2108873\u003c/li\u003e\n\u003cli\u003eWittner BS, Sgroi DC, Ryan PD, et al (2008) Analysis of the MammaPrint Breast Cancer Assay in a Predominantly Postmenopausal Cohort. Clin Cancer Res 14:2988\u0026ndash;2993. https://doi.org/10.1158/1078-0432.CCR-07-4723\u003c/li\u003e\n\u003cli\u003eWishart GC, Bajdik CD, Azzato EM, et al (2011) A population-based validation of the prognostic model PREDICT for early breast cancer. European Journal of Surgical Oncology (EJSO) 37:411\u0026ndash;417. https://doi.org/10.1016/j.ejso.2011.02.001\u003c/li\u003e\n\u003cli\u003eHoskins KF, Danciu OC, Ko NY, Calip GS (2021) Association of Race/Ethnicity and the 21-Gene Recurrence Score With Breast Cancer\u0026ndash;Specific Mortality Among US Women. JAMA Oncology 7:370\u0026ndash;378. https://doi.org/10.1001/jamaoncol.2020.7320\u003c/li\u003e\n\u003cli\u003eCrager M, Wijayawardana SR, Gruver AM, et al (2022) Population-based estimate for the correlation of the Oncotype Dx Breast Recurrence Score\u0026reg; result and Ki-67 IHC MIB-1 pharmDx in HR+, HER2\u0026minus;, node-positive early breast cancer. Breast Cancer Res 24:74. https://doi.org/10.1186/s13058-022-01571-7\u003c/li\u003e\n\u003cli\u003eBattisti NML, Reed MWR, Herbert E, et al (2021) Bridging the Age Gap in breast cancer: Impact of chemotherapy on quality of life in older women with early breast cancer. European Journal of Cancer 144:269\u0026ndash;280. https://doi.org/10.1016/j.ejca.2020.11.022\u003c/li\u003e\n\u003cli\u003eWalsh DJ, Sahm LJ, O\u0026rsquo;Driscoll M, et al (2023) Hospitalization due to adverse drug events in older adults with cancer: A retrospective analysis. J Geriatr Oncol 14:101540. https://doi.org/10.1016/j.jgo.2023.101540\u003c/li\u003e\n\u003cli\u003eMagnuson A, Sedrak MS, Gross CP, et al (2021) Development and Validation of a Risk Tool for Predicting Severe Toxicity in Older Adults Receiving Chemotherapy for Early-Stage Breast Cancer. J Clin Oncol 39:608\u0026ndash;618. https://doi.org/10.1200/JCO.20.02063\u003c/li\u003e\n\u003cli\u003eRamspek CL, Jager KJ, Dekker FW, et al (2020) External validation of prognostic models: what, why, how, when and where? Clin Kidney J 14:49\u0026ndash;58. https://doi.org/10.1093/ckj/sfaa188\u003c/li\u003e\n\u003cli\u003eSiontis GCM, Tzoulaki I, Castaldi PJ, Ioannidis JPA (2015) External validation of new risk prediction models is infrequent and reveals worse prognostic discrimination. J Clin Epidemiol 68:25\u0026ndash;34. https://doi.org/10.1016/j.jclinepi.2014.09.007\u003c/li\u003e\n\u003cli\u003eDebray TPA, Vergouwe Y, Koffijberg H, et al (2015) A new framework to enhance the interpretation of external validation studies of clinical prediction models. J Clin Epidemiol 68:279\u0026ndash;289. https://doi.org/10.1016/j.jclinepi.2014.06.018\u003c/li\u003e\n\u003cli\u003eEdge SB, Compton CC (2010) The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol 17:1471\u0026ndash;1474. https://doi.org/10.1245/s10434-010-0985-4\u003c/li\u003e\n\u003cli\u003eCommon Terminology Criteria for Adverse Events (CTCAE). 155\u003c/li\u003e\n\u003cli\u003eRiedel F, Hoffmann AS, Moderow M, et al (2020) Time trends of neoadjuvant chemotherapy for early breast cancer. International Journal of Cancer 147:3049\u0026ndash;3058. https://doi.org/10.1002/ijc.33122\u003c/li\u003e\n\u003cli\u003eMasuda N, Lee S-J, Ohtani S, et al (2017) Adjuvant Capecitabine for Breast Cancer after Preoperative Chemotherapy. New England Journal of Medicine 376:2147\u0026ndash;2159. https://doi.org/10.1056/NEJMoa1612645\u003c/li\u003e\n\u003cli\u003evon Minckwitz G, Huang C-S, Mano MS, et al (2019) Trastuzumab Emtansine for Residual Invasive HER2-Positive Breast Cancer. New England Journal of Medicine 380:617\u0026ndash;628. https://doi.org/10.1056/NEJMoa1814017\u003c/li\u003e\n\u003cli\u003eBaskin AS, Huppert LA, Kelil T, et al (2024) The neoadjuvant approach to treatment of breast cancer: Multidisciplinary management to improve outcomes. Surgical Oncology Insight 1:100059. https://doi.org/10.1016/j.soi.2024.100059\u003c/li\u003e\n\u003cli\u003eUchiyama M, Miyazaki M, Hayashi T, et al (2024) Assessing the ability of the Cancer and Aging Research Group tool to predict chemotherapy toxicity in older Japanese patients: A prospective observational study. Journal of Geriatric Oncology 15:101814. https://doi.org/10.1016/j.jgo.2024.101814\u003c/li\u003e\n\u003cli\u003ePang A, Jiali L, Ng A, et al (2022) Use of the Cancer and Aging Research Group Predictive Model for Chemotherapy-Related Toxic Effects in a Multiethnic, Older Adult Asian Population. JAMA Netw Open 5:e2237196. https://doi.org/10.1001/jamanetworkopen.2022.37196\u003c/li\u003e\n\u003cli\u003eOstwal V, Ramaswamy A, Bhargava P, et al (2021) Cancer Aging Research Group (CARG) score in older adults undergoing curative intent chemotherapy: a prospective cohort study. BMJ Open 11:e047376. https://doi.org/10.1136/bmjopen-2020-047376\u003c/li\u003e\n\u003cli\u003eHurria A, Togawa K, Mohile SG, et al (2011) Predicting Chemotherapy Toxicity in Older Adults With Cancer: A Prospective Multicenter Study. JCO 29:3457\u0026ndash;3465. https://doi.org/10.1200/JCO.2011.34.7625\u003c/li\u003e\n\u003cli\u003eLingamaiah D, Bushi G, Gaidhane S, et al (2025) Falls among geriatric cancer patients: a systematic review and meta-analysis of prevalence and risk across cancer types. BMC Geriatr 25:179. https://doi.org/10.1186/s12877-025-05722-1\u003c/li\u003e\n\u003cli\u003eBaxter MA, Rowe M, Zucker K, et al (2024) UK national observational cohort study investigating Tolerance of Anti-cancer Systemic Therapy in the Elderly: the TOASTIE study. bmjonc 3:. https://doi.org/10.1136/bmjonc-2024-000459\u003c/li\u003e\n\u003cli\u003eMbewe A, Pike P, Lewis R, et al (2021) Implementing the cancer and aging research group (CARG) tool in the ambulatory oncology setting to drive informed treatment selection. JCO 39:209\u0026ndash;209. https://doi.org/10.1200/JCO.2020.39.28_suppl.209\u003c/li\u003e\n\u003cli\u003eHamaker ME, Wildes TM, Rostoft S (2017) Time to Stop Saying Geriatric Assessment Is Too Time Consuming. J Clin Oncol 35:2871\u0026ndash;2874. https://doi.org/10.1200/JCO.2017.72.8170\u003c/li\u003e\n\u003cli\u003eMintchev ME, Kalra AG, Kou C-TJ, et al Use of a Chemotherapy Toxicity Prediction Tool to Decrease Risks for Hospitalization in Older Patients. Cureus 14:e24465. https://doi.org/10.7759/cureus.24465\u003c/li\u003e\n\u003cli\u003eMohile SG, Dale W, Somerfield MR, et al (2018) Practical Assessment and Management of Vulnerabilities in Older Patients Receiving Chemotherapy: ASCO Guideline for Geriatric Oncology. J Clin Oncol 36:2326\u0026ndash;2347. https://doi.org/10.1200/JCO.2018.78.8687\u003c/li\u003e\n\u003cli\u003eGrim S, Kotz A, Kotz G, et al (2024) Development and validation of electronic health record-based, machine learning algorithms to predict quality of life among family practice patients. Sci Rep 14:30077. https://doi.org/10.1038/s41598-024-80064-3\u003c/li\u003e\n\u003cli\u003eSteele AJ, Denaxas SC, Shah AD, et al (2018) Machine learning models in electronic health records can outperform conventional survival models for predicting patient mortality in coronary artery disease. PLoS One 13:e0202344. https://doi.org/10.1371/journal.pone.0202344\u003c/li\u003e\n\u003cli\u003ePeterson DJ, Ostberg NP, Blayney DW, et al (2021) Machine Learning Applied to Electronic Health Records: Identification of Chemotherapy Patients at High Risk for Preventable Emergency Department Visits and Hospital Admissions. JCO Clin Cancer Inform 1106\u0026ndash;1126. https://doi.org/10.1200/CCI.21.00116\u003c/li\u003e\n\u003cli\u003eWang Z, Wang T, Xie Y, et al (2025) Unraveling the role of adjuvant chemotherapy in elderly triple-negative breast cancer: Insights from competing risk analysis using SEER data. Cancer Epidemiology 97:102853. https://doi.org/10.1016/j.canep.2025.102853\u003c/li\u003e\n\u003cli\u003eHurria A, Dale W, Mooney M, et al (2014) Designing Therapeutic Clinical Trials for Older and Frail Adults With Cancer: U13 Conference Recommendations. J Clin Oncol 32:2587\u0026ndash;2594. https://doi.org/10.1200/JCO.2013.55.0418\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"breast-cancer-research-and-treatment","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"brea","sideBox":"Learn more about [Breast Cancer Research and Treatment](https://www.springer.com/journal/10549)","snPcode":"10549","submissionUrl":"https://submission.nature.com/new-submission/10549/3","title":"Breast Cancer Research and Treatment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"breast cancer, older adults, geriatric oncology, chemotherapy, toxicity, external validation","lastPublishedDoi":"10.21203/rs.3.rs-7468335/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7468335/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003ePurpose:\u003c/strong\u003e Decision-making for chemotherapy in early breast cancer (EBC) in OA (older adults: age ≥65 years) is complex due to frailty, multimorbidity, and competing risks for mortality. Magnuson (2021) developed a chemotherapy toxicity prediction score, CARG-BC; its external validation can improve generalizability.\u003c/p\u003e\n\u003cp\u003eObjectives: CARG-BC’s ability to predict grade 3+ chemotoxicity in OA with EBC (primary), unplanned healthcare use, and changes to chemotherapy protocol (secondary).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e A single centre retrospective cohort study comprising OA with EBC who received (neo)adjuvant chemotherapy from 2013-2023. Clinical, demographic, CARG-BC, and healthcare usage variables were extracted from patient records. Risk groups based on CARG-BC score were compared using T-test (continuous variables) \u0026amp; χ2 test (categorical variables). Toxicity risk based on CARG-BC score was assessed using logistic regression. The predictive ability of the CARG-BC score was evaluated by calculating AUC.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eOf 243 patients, the median age was 70 years (range 65-86), 99.6% female, 80.2% with comorbidities, 33.7% with polypharmacy, 28.8% living alone, and 8.2% seen in the geriatric oncology clinic. Over half (53.9%) had grade 3+ toxicities. Healthcare utilization included 19.8% of patients with at least one unplanned clinic visit, 29.6% an emergency care visit, and 14.4% a hospitalization. The median CARG-BC score was 7 (IQR 3, 8) and the CARG-BC AUC was 0.76 (95% Confidence interval [CI] 0.70, 0.82). The odds of grade 3+ toxicity is increased by 1.33 times per CARG-BC point increase.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e The CARG-BC model retained good discrimination for grade ≥3 chemotoxicity and should be used in shared-decision making with OA.\u003c/p\u003e","manuscriptTitle":"Using the Cancer Aging and Research Group- Breast Cancer (CARG-BC) predictive model in older adults (OA) with early breast cancer: an external validation study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-22 10:05:25","doi":"10.21203/rs.3.rs-7468335/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-29T19:48:15+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-29T11:17:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"142468292926481074957604178915398277609","date":"2025-10-13T18:56:58+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-12T21:06:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"106609226641759858457258560015725434388","date":"2025-09-13T22:24:06+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-11T22:22:04+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-28T01:38:34+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-28T01:37:49+00:00","index":"","fulltext":""},{"type":"submitted","content":"Breast Cancer Research and Treatment","date":"2025-08-27T06:14:29+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"breast-cancer-research-and-treatment","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"brea","sideBox":"Learn more about [Breast Cancer Research and Treatment](https://www.springer.com/journal/10549)","snPcode":"10549","submissionUrl":"https://submission.nature.com/new-submission/10549/3","title":"Breast Cancer Research and Treatment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"00607e86-b08e-493e-9238-7cecc2eef82e","owner":[],"postedDate":"September 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-08T16:05:36+00:00","versionOfRecord":{"articleIdentity":"rs-7468335","link":"https://doi.org/10.1007/s10549-025-07869-2","journal":{"identity":"breast-cancer-research-and-treatment","isVorOnly":false,"title":"Breast Cancer Research and Treatment"},"publishedOn":"2025-12-02 15:57:39","publishedOnDateReadable":"December 2nd, 2025"},"versionCreatedAt":"2025-09-22 10:05:25","video":"","vorDoi":"10.1007/s10549-025-07869-2","vorDoiUrl":"https://doi.org/10.1007/s10549-025-07869-2","workflowStages":[]},"version":"v1","identity":"rs-7468335","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7468335","identity":"rs-7468335","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}