Machine Learning May Predict Patient Satisfaction After Breast Reconstruction

In In The News by Barbara Jacoby

By: Julia Faith Sklar

From: cancertherapyadvisor.com

Machine learning increasingly supports physician decisions by making it easier to detect patterns in data as a means of predicting patient outcomes. In breast cancer, that now could apply to every stage of the experience, from diagnostics to mastectomy and breast reconstruction. At the annual meeting of the American Society of Clinical Oncology — which was virtual this year, due to the ongoing coronavirus pandemic — a consortium of researchers presented an abstract detailing how machine learning algorithms were able to correctly predict how individual patients would feel about their breast reconstruction.1 Using this tool in a clinical setting could help physicians guide patients through the recovery process in a way that better anticipates, and subsequently supports, their emotional reaction to this intensely personal medical procedure.

Physician-researchers across 11 institutions in the United States and Canada trained 4 different types of machine learning algorithms — regularized regression, Support Vector Machine, Neural Network, Regression Tree — to predict with 95% accuracy whether a specific patient would be satisfied or dissatisfied with their breast reconstruction 2 years after their operation.

The algorithms were trained on data collected during a 5-year clinical trial of 4436 patients that assessed patient-reported outcomes in response to different techniques in breast reconstruction — crucially, it included both patients who underwent reconstruction jointly with mastectomy and those who held off on a secondary, postmastectomy reconstruction procedure until a later date.

Allowing machine learning to pick up where the clinical trial left off is the next step in the trial’s mission: “In order to actively participate in the reconstruction decision-making process, breast cancer survivors need objective, up-to-date information on breast reconstruction outcomes from the patient’s perspective.”

Increasingly, aiming for breast conservation is the gold standard in primary therapy for breast cancer, but for those who ultimately must undergo mastectomy, the postmastectomy options can be hard for patients to wade through. The choices are multifold. One option for many patients, of course, is to stop at the point of mastectomy. Not every patient medically requires a breast reconstruction; some instead choose to wear bras outfitted with breast prostheses, while others go without them altogether.

For patients who do consider breast reconstruction critical to their psychological well-being, sense of self, and quality of life, however, a second level of choices abounds for which reconstruction technique will yield the best results for them: expander/implant, latissimus dorsi/implant (LD), pedicle transverse rectus abdominis musculocutaneous (PTRAM), free TRAM (FTRAM), deep inferior epigastric perforator (DIEP), superficial inferior epigastric artery (SIEA), superior gluteal artery perforator (SGAP), or inferior gluteal artery perforator (IGAP).

In the clinical trial, patient-reported outcomes around complications, postoperative pain, psychosocial well-being, physical functioning, fatigue, patient satisfaction and costs were assessed at 1 week, 3 months, 1 year, and 2 years postoperatively. The data set then used to train the algorithms was merely patient satisfaction after 2 years.

One major ongoing shortcoming of machine learning algorithms in many fields is that they are trained on data from primarily white populations. In the original clinical trial, however, race and ethnicity were key data points collected from the cohort and tracked with outcomes, which is rare, and means the subsequent machine learning algorithms presented in the abstract are less likely to suffer from racial and ethnic biases — a foundational need for the development of clinical tools.

In other clinical breast cancer settings, machine learning is also being tested as a tool to support better patient outcomes.2 Researchers in Switzerland have been developing an algorithmic tool to help physicians make data-driven, personalized decisions around which patients would most benefit from preventive treatments. The algorithm helps identify and target high-risk patients, while helping lower-risk patients avoid unnecessary prevention that could be costly, painful, and upsetting to go through. This is machine learning applied at the earliest, most cautious point of breast cancer care: prevention. The abstract presented at the ASCO20 meeting bookends the field by offering an algorithm at 1 of the final points of care: reconstruction.

Ultimately, what the algorithms show is that the best outcome predictor is baseline satisfaction with breasts. But other informative predictors are whether the patient experienced “radiation during or after reconstruction, nipple-sparing and mixed mastectomy, implant-based reconstruction, chemotherapy, unilateral mastectomy, lower psychological well-being, and obesity,” according to the poster.

Group-level guidance often underpins clinical decisions around mastectomy and reconstruction when such a personal decision deserves a more personalized tool to figure out whether reconstruction is the right choice, and, if so, which reconstruction technique will result in the highest satisfaction for the patient.

References

  1. Pfob A, Mehrara B, Nelson J, Wilkins E, Pusic A, Sidey-Gibbons C. Towards data-driven decision-making for breast cancer patients undergoing mastectomy and reconstruction: Prediction of individual patient-reported outcomes at two-year follow-up using machine learning. Presented at: ASCO20 Virtual Scientific Program. J Clin Oncol. 38;2020(suppl):abstr 520.
  2. Ming C, Viassolo V, Probst-Hensch N, Chappuis PO, Dinov ID, Katapodi MC. Machine learning techniques for personalized breast cancer risk prediction: comparison with the BCRAT and BOADICEA models. Breast Cancer Res. 2019;21(7):75.