Can Simple Clinical Factors Predict Metastatic Potential of Neuroendocrine Tumors?

Researchers have developed a clinical prediction model for the metastatic potential of pheochromocytoma and paraganglioma (PPGL), incorporating age, tumor size, extra-adrenal location, and norepinephrine-secretory type. Other risk prediction models have incorporated some similar information in the past.

“Unlike other cancers, malignant PPGL can be diagnosed only after the documentation of metastases,” wrote study authors led by Yoon Young Cho, MD, of Gyeongsang National University School of Medicine in Jinju, South Korea. “Early detection of PPGL with metastatic potential is crucial for better prognosis … therefore, attempts have been made to find histologic or clinical markers that can predict metastatic potential.”

Though various clinical and histologic markers have been proposed, the authors wrote that no multifactorial scoring system using preoperative clinical risk factors has been developed. They developed such a system called the ASES score based on 333 patients who underwent resection for PPGL at 2 medical centers; the patients were followed for a median of 27.5 months after surgery. The results were published in the September issue of Surgery.

In total, metastases occurred in 23 of the 333 patients (6.9%). On a univariate analysis, metastatic disease was associated with several of the factors studied, including age of 35 years or younger (hazard ratio [HR], 2.74; 95% CI, 1.19–6.35; P = .019), tumor size of 6.0 cm or larger (HR, 2.43; 95% CI, 1.06–5.56; P = .036), extra-adrenal location (HR, 2.73; 95% CI, 1.10–7.40; P = .048), and tumors producing only norepinephrine (HR, 2.96; 95% CI, 1.30–6.76; P = .015).

Each of these variables was assigned a point score of either 1-when the factor was present-or 0, when it was not. The area under the curve for this score as a predictor of metastatic potential was 0.735; at a cutoff point of a score of 2, it had a sensitivity of 60.9%, a specificity of 80.0%, and a negative predictive value of 96.5%. Those with a score of 2 or higher had a 5-year metastasis-free survival rate of 74.5% compared with 94.7% in those with scores below 2; at 10 years, these rates were 29.6% and 86.3%, respectively (P < .0001).

In an analysis of the 305 patients with pheochromocytoma, the cutoff of 2 again produced a significant difference in metastasis-free survival.

“The excellent negative predictive value of the ASES score indicates that this prediction model can identify PPGL patients with low metastatic potential,” wrote the authors.

Arthur S. Tischler, MD, an expert on neuroendocrine tumors at Tufts Medical Center in Boston, who was not involved with this research, told Cancer Network that this scoring system is “just a way of packaging information that is already well known.” Size of tumor and extra-adrenal location are already included in American Joint Committee on Cancer staging guidelines, he said, and younger patients often have a mutation in the SDHB gene, which alone is a strong risk factor for metastasis. He noted that a previously developed system known as GAPP also includes some of these parameters.

The authors, though, note some limitations of GAPP and other models, and wrote that the new model involves easily obtained clinical factors. “Using four simple clinical parameters, this system can help clinicians distinguish patients at low risk for metastases,” they concluded. “The ASES score is a practical tool that can be used in conjunction with a histologic scoring system.”