Treatment outcome in patients with stage III breast cancer treated with neoadjuvant chemotherapy

Does the Type of Reconstruction Matter? A Propensity Score Analysis of Immediate Postmastectomy Implant and Flap Reconstruction

BACKGROUND

No randomized controlled trials have compared implant and flap reconstruction. Recently, worse longitudinal outcomes have been suggested for flap reconstruction. The authors compared long-term oncologic outcomes of postmastectomy breast reconstruction using propensity score matching.

METHODS

A retrospective study of postmastectomy reconstruction was achieved using the Weill Cornell Breast Cancer Registry between 1998 and 2019. Patients were matched using propensity scores based on demographic, clinical, and surgical characteristics. Kaplan-Meier estimates, Cox-regression models, and restricted mean survival times (RMST) were used to evaluate patient outcomes.

RESULTS

Before matching, 1395 implant and 586 flap patients were analyzed. No difference in overall survival and recurrence were observed. Multivariable models showed decreased survival for Medicare/Medicaid [hazard ratio (HR), 3.09; 95% CI, 1.63 to 5.87; P < 0.001], pathologic stage II (HR, 2.98; 95% CI, 1.12 to 7.90; P = 0.028), stage III (HR, 4.88; 95% CI, 1.54 to 15.5; P = 0.007), 11 to 20 lymph nodes positive (HR, 3.66; 95% CI, 1.31 to 10.2; P = 0.013), more than 20 lymph nodes positive (HR, 6.41; 95% CI, 1.49 to 27.6; P = 0.013). RMST at 10 years after flap reconstruction showed 2 months of decreased survival time compared with implants (9.56 versus 9.74 years; 95% CI, -0.339 to -0.024; P = 0.024). After matching, 563 implant and 563 flap patients were compared. Reconstruction was not associated with overall survival and recurrence. RMST between implant and flap reconstruction showed no difference in each 5-year interval over 20 years.

CONCLUSION

Postmastectomy breast reconstruction was not associated with a difference in long-term oncologic outcomes over a 20-year period.

CLINICAL QUESTION/LEVEL OF EVIDENCE

Therapeutic, III.

Outcome-Oriented Deep Temporal Phenotyping of Disease Progression

Chronic diseases evolve slowly throughout a patient's lifetime creating heterogeneous progression patterns that make clinical outcomes remarkably varied across individual patients. A tool capable of identifying temporal phenotypes based on the patients different progression patterns and clinical outcomes would allow clinicians to better forecast disease progression by recognizing a group of similar past patients, and to better design treatment guidelines that are tailored to specific phenotypes. To build such a tool, we propose a deep learning approach, which we refer to as outcome-oriented deep temporal phenotyping (ODTP), to identify temporal phenotypes of disease progression considering what type of clinical outcomes will occur and when based on the longitudinal observations. More specifically, we model clinical outcomes throughout a patient's longitudinal observations via time-to-event (TTE) processes whose conditional intensity functions are estimated as non-linear functions using a recurrent neural network. Temporal phenotyping of disease progression is carried out by our novel loss function that is specifically designed to learn discrete latent representations that best characterize the underlying TTE processes. The key insight here is that learning such discrete representations groups progression patterns considering the similarity in expected clinical outcomes, and thus naturally provides outcome-oriented temporal phenotypes. We demonstrate the power of ODTP by applying it to a real-world heterogeneous cohort of 11 779 stage III breast cancer patients from the U.K. National Cancer Registration and Analysis Service. The experiments show that ODTP identifies temporal phenotypes that are strongly associated with the future clinical outcomes and achieves significant gain on the homogeneity and heterogeneity measures over existing methods. Furthermore, we are able to identify the key driving factors that lead to transitions between phenotypes which can be translated into actionable information to support better clinical decision-making.