Study Challenges Necessity of Sentinel Lymph Node Biopsy in Breast Cancer Patients

In a recent study published in JCO Clinical Cancer Informatics, researchers from the University of Pittsburgh and UPMC Hillman Cancer Center have presented compelling evidence that sentinel lymph node biopsy (SLNB) may do more harm than good for middle-aged women with estrogen receptor-positive (ER+) breast cancer.

The study suggests extending current clinical guidelines that recommend de-escalating surgery for women over 70 to include post-menopausal patients aged 55 and older.

Using a novel artificial intelligence (AI) pipeline developed by Realyze Intelligence, a UPMC Enterprises portfolio company, the researchers analyzed electronic health records (EHRs) of 925 patients aged 55 or older diagnosed with ER+ breast cancer between 2015 and 2017. The findings indicate that the risks associated with SLNB, including lymphedema, may not justify the procedure's benefits for this patient group.

The Procedure and Its Risks

Sentinel lymph node biopsy involves the surgical removal of one or more lymph nodes that drain from a patient's primary tumor, allowing oncologists to detect whether the cancer has spread. Despite being a standard procedure for breast cancer patients, SLNB carries risks, including lymphedema—a condition characterized by the buildup of lymphatic fluid causing swelling, discomfort, and mobility issues.

"Even though the risk of lymphedema is relatively low, it can have devastating effects on a person's quality of life," said Dr. Emilia Diego, senior author of the study, associate professor of surgery, and chief of the Breast Surgery Division at Pitt. "As medical professionals, we must continuously seek ways to improve care while minimizing treatment's untoward outcomes. This study suggests that SLNB may not be necessary in every breast cancer case."

Analyzing the Data

The AI pipeline, utilizing natural language understanding (NLU), filtered through hundreds of thousands of clinical notes to extract valuable data not routinely documented in cancer records. The tool was 95% accurate in identifying cases of breast or arm lymphedema, highlighting its efficacy in extracting critical information from unstructured data.

The study found that only 13% of patients aged 55-69 and 7% of those aged 70 or older had positive sentinel lymph node biopsies, indicating a low risk of metastasis. For patients with lymph node involvement, the majority had just 1-3 positive nodes—a result that does not significantly impact treatment decisions without additional genetic data. Notably, lymphedema rates were higher than rates of lymph node involvement.

Implications for Clinical Practice

These findings build on previous research showing that women over 70 with early-stage ER+ breast cancer have very low rates of metastasis, leading to efforts to de-escalate treatment for this group. In 2016, the Society of Surgical Oncology recommended against routine SLNB for ER+ breast cancer patients aged over 70 with tumors less than 2cm in diameter.

"There's a lot of valuable data in the EHR that is hidden within unstructured physician notes," said Neil Carleton, lead author of the study and a graduate student in Pitt's Medical Scientist Training Program. "Realyze's power lies in pulling out information by analyzing the language and context of those notes."

Future Directions

Given the study's findings, the researchers advocate for updating clinical guidelines to include women aged 55 and older with small, ER+ tumors in the recommendations for de-escalating SLNB. This change could spare many women the potential harms of unnecessary surgery.

"Our study, along with recently published clinical trial data, suggests that recommendations aimed at de-escalating surgery could be extended to younger women with small, ER+ disease," said Carleton.

This research, supported by the National Cancer Institute, the Breast Cancer Research Foundation, and Susan G. Komen, marks a significant step towards personalized, less invasive cancer care, emphasizing treatment tailored to each patient's specific needs.