Machine Learning Approach May Help Identify Melanoma Survivors Who Face a High Risk of Cancer Recurrence

Researchers have developed and validated a machine learning-based method to predict which patients with early-stage melanoma are most likely to experience a cancer recurrence.

The tool, described in npj Precision Oncology, may help identify patients who would benefit from aggressive treatment even at early stages of melanoma.

Most patients with early-stage melanoma are treated with surgery to remove cancerous cells, but patients with more advanced cancer often receive immune checkpoint inhibitors, which effectively strengthen the immune response against tumor cells but also carry significant side effects.

"There is an urgent need to develop predictive tools to assist in the selection of high-risk patients for whom the benefits of immune checkpoint inhibitors would justify the high rate of morbid and potentially fatal immunologic adverse events observed with this therapeutic class," said senior author Yevgeniy R. Semenov, MD, Massachusetts General Hospital, Boston, Massachusetts. "Reliable prediction of melanoma recurrence can enable more precise treatment selection for immunotherapy, reduce progression to metastatic disease and improve melanoma survival while minimising exposure to treatment toxicities."

To help achieve this, Dr. Semenov and colleagues assessed the effectiveness of algorithms based on machine learning that used data from patient electronic health records to predict melanoma recurrence. Specifically, the team collected data from 1,720 patients with early-stage melanomas (1,172 from MassGen and 548 from Dana-Farber Cancer Institute in Boston) and extracted 36 clinical and pathologic features of these cancers from electronic health records to predict patients' recurrence risk with machine learning algorithms. Models were evaluated internally and externally.

In the internal and external validations, respectively, the model achieved a recurrence classification performance of area under the curve (AUC) 0.845 and 0.812, and a time-to-event prediction performance of time-dependent AUC of 0.853 and 0.820. Breslow tumour thickness and mitotic rate were identified as the most predictive features.

"Our comprehensive risk prediction platform using novel machine learning approaches to determine the risk of early-stage melanoma recurrence reached high levels of classification and time to event prediction accuracy," says Semenov. "Our results suggest that machine learning algorithms can extract predictive signals from clinicopathologic features for early-stage melanoma recurrence prediction, which will enable the identification of patients who may benefit from adjuvant immunotherapy."

Reference:https://www.nature.com/articles/s41698-022-00321-4

SOURCE: Massachusetts General Hospital

Sent from my iPhone

Benjamin Hidalgo-Matlock
Skin Care Physicians of Costa Rica

Clinica Victoria en San Pedro: 4000-1054
Momentum Escazu: 2101-9574

Please excuse the shortness of this message, as it has been sent from
a mobile device.