Dataset Information

Machine learning to predict individual patient-reported outcomes at 2-year follow-up for women undergoing cancer-related mastectomy and breast reconstruction (INSPiRED-001).

ABSTRACT:

Background

Women undergoing cancer-related mastectomy and reconstruction are facing multiple treatment choices where post-surgical satisfaction with breasts is a key outcome. We developed and validated machine learning algorithms to predict patient-reported satisfaction with breasts at 2-year follow-up to better inform the decision-making process for women with breast cancer.

Methods

We trained, tested, and validated three machine learning algorithms (logistic regression (LR) with elastic net penalty, Extreme Gradient Boosting (XGBoost) tree, and neural network) to predict clinically important differences in satisfaction with breasts at 2-year follow-up using the validated BREAST-Q. We used data from 1553 women undergoing cancer-related mastectomy and reconstruction who were followed-up for two years at eleven study sites in North America from 2011 to 2016. 10-fold cross-validation was used to train and test the algorithms on data from 10 of the 11 sites which were further validated using the additional site's data. Area-under-the-receiver-operating-characteristics-curve (AUC) was the primary outcome measure.

Results

Of 1553 women, 702 (45.2%) experienced an improved satisfaction with breasts and 422 (27.2%) a decreased satisfaction. In the validation set (n = 221), the algorithms showed equally high performance to predict improved or decreased satisfaction with breasts (all P > 0.05): For improved satisfaction AUCs were 0.86-0.87 and for decreased satisfaction AUCs were 0.84-0.85.

Conclusion

Long-term, individual patient-reported outcomes for women undergoing mastectomy and breast reconstruction can be accurately predicted using machine learning algorithms. Our algorithms may be used to better inform clinical treatment decisions for these patients by providing accurate estimates of expected quality of life.

SUBMITTER: Pfob A

PROVIDER: S-EPMC8551470 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Purpose The goals of immediate postmastectomy breast reconstruction are to minimize deformity and optimize quality of life as perceived by patients. We prospectively evaluated patient-reported outcomes (PROs) in women undergoing immediate implant-based or autologous reconstruction. Methods Women undergoing immediate postmastectomy reconstruction for invasive cancer and/or carcinoma in situ were enrolled at 11 sites. Women underwent implant-based or autologous tissue reconstruction. Patients completed the BREAST-Q, a condition-specific PRO measure for breast surgery patients, and Patient-Reported Outcomes Measurement Information System-29, a generic PRO measure, before and 1 year after surgery. Mean changes in PRO scores were summarized. Mixed-effects regression models were used to compare PRO scores across procedure types. Results In total, 1,632 patients (n = 1,139 implant, n = 493 autologous) were included; 1,183 (72.5%) responded to 1-year questionnaires. After analysis was controlled for baseline values, patients who underwent autologous reconstruction had greater satisfaction with their breasts than those who underwent implant-based reconstruction (difference, 6.3; P < .001), greater sexual well-being (difference, 4.5; P = .003), and greater psychosocial well-being (difference, 3.7; P = .02) at 1 year. Patients in the autologous reconstruction group had improved satisfaction with breasts (difference, 8.0; P = .002) and psychosocial well-being (difference, 4.6; P = .047) compared with preoperative baseline. Physical well-being of the chest was not fully restored in either the implant group (difference, -3.8; P = .001) or autologous group (-2.2; P = .04), nor was physical well-being of the abdomen in patients who underwent autologous reconstruction (-13.4; P < .001). Anxiety and depression were mitigated at 1 year in both groups. Compared with their baseline reports, patients who underwent implant reconstruction had decreased fatigue (difference, -1.4; P = .035), whereas patients who underwent autologous reconstruction had increased pain interference (difference, 2.0; P = .006). Conclusion At 1 year after mastectomy, patients who underwent autologous reconstruction were more satisfied with their breasts and had greater psychosocial and sexual well-being than those who underwent implant reconstruction. Although satisfaction with breasts was equal to or greater than baseline levels, physical well-being was not fully restored. This information can help patients better understand expected outcomes and may guide innovations to improve outcomes.

Project description:Neoadjuvant chemotherapy (NC) is increasingly being used in patients with breast cancer, and evidence-based reports related to its independent effects on morbidity after mastectomy with immediate breast reconstruction are limited.To determine the effect of NC on 30-day postoperative morbidity in women undergoing mastectomy with or without immediate breast reconstruction.All women undergoing mastectomy with or without immediate breast reconstruction from January 1, 2005, through December 31, 2011, at university and private hospitals internationally were analyzed using the American College of Surgeons National Surgical Quality Improvement Program 2005-2011 databases. Patients who received NC were compared with those without a history of NC to estimate the relative odds of 30-day postoperative overall, systemic, and surgical site morbidity using model-wise multivariable logistic regression.Neoadjuvant chemotherapy.Thirty-day postoperative morbidity (overall, systemic, and surgical site).Of 85,851 women, 66,593 (77.6%) underwent mastectomy without breast reconstruction, with 2876 (4.3%) receiving NC; 7893 patients were excluded because of missing exposure data. The immediate breast reconstruction population included 19,258 patients (22.4%), with 820 (4.3%) receiving NC. After univariable analysis, NC was associated with a 20% lower odds of overall morbidity in the group undergoing mastectomy without breast reconstruction (odds ratio [OR], 0.80; 95% CI, 0.71-0.91) but had no significant effect in the immediate breast reconstruction group (OR, 0.98; 95% CI, 0.79-1.23). After adjustment for confounding, NC was independently associated with lower overall morbidity in the group undergoing mastectomy without breast reconstruction (OR, 0.61; 95% CI, 0.51-0.73) and the immediate tissue expander reconstruction subgroup (OR, 0.49; 95% CI, 0.30-0.84). Neoadjuvant chemotherapy was associated with decreased odds of systemic morbidity in 4 different populations: complete sample (OR, 0.59; 95% CI, 0.49-0.71), mastectomy without breast reconstruction (OR, 0.59; 95% CI, 0.48-0.72), any immediate breast reconstruction (OR, 0.57; 95% CI, 0.37-0.88), and the tissue expander subgroup (OR, 0.41; 95% CI, 0.23-0.72).Our study supports the safety of NC in women undergoing mastectomy with or without immediate breast reconstruction. Neoadjuvant chemotherapy is associated with lower overall morbidity in the patients undergoing mastectomy without breast reconstruction and in those undergoing tissue expander breast reconstruction. In addition, the odds of systemic morbidity were decreased in patients undergoing mastectomy with and without immediate breast reconstruction. The mechanisms behind the protective association of NC remain unknown and warrant further investigation.

Project description:BackgroundAfter breast cancer, some patients report residual pain-related upper limb disability without physical impairment. Although pain and altered proprioception are known to affect the working body schema (WBS), there is little available evidence investigating the WBS of breast cancer survivors (BrCS). WBS-body representations in the brain-affect the "neuromatrix" that modulates pain sensitivity and the threshold for threatening stimuli. The aim of this study was to investigate whether WBS was disrupted after mastectomy with immediate breast reconstruction (IBR) for breast cancer and whether pain and proprioceptive changes affected WBS.MethodsThirty-five BrCS participated in the 4-month follow-up study. They were observed at 1 and 4 months postoperatively. The main outcome measures were the left right judgement test (LRJT) results, absolute angle error, pectoralis minor length index (PMI), pain, and Quick-Disabilities of the Arm, Shoulder, and Hand (Q-DASH) score. They were measured at each observation, and parametric tests were performed to identify the nature of WBS.ResultsBoth the reaction time and accuracy of the hand LRJT were poorer than those of the foot and back LRJT (p < 0.001). The hand LRJT reaction time and accuracy were unchanged over the total follow-up period (p = 0.77 and p = 0.47, respectively). There was a weak correlation between the LRJT reaction time and PMI (r = -0.26, p = 0.07), pain severity (r = 0.37, p = 0.02), and Q-DASH score (r = 0.37, p = 0.02). There was also a weak correlation between LRJT accuracy and Q-DASH score (r = -0.31, p = 0.04). The LRJT accuracy of BrCS who underwent surgery on their dominant side was higher than that of BrCS who underwent surgery on their non-dominant side (p = 0.002). Regression analysis found a weak but significant relationship between the early hand LRJT results and late pain severity (adjusted R2 = 0.179, p = 0.007). A similar relationship was found between early hand LRJT results and Q-DASH score (adjusted R2 = 0.099, p = 0.039).ConclusionTo the best of our knowledge, this is the first study providing the nature of WBS after mastectomy with IBR. In this population, it is necessary to postoperatively preserve WBS integrity for pain and upper limb disability.