84th Annual Meeting of the American Thyroid Association:
Medullary Thyroid Cancer Nomogram
Ian Ganly, MD, PhD presented findings from a medullary thyroid cancer clinical study conducted by Memorial Sloan-Kettering Cancer Center (MSKCC). First, Dr. Ganly reviewed some facts about medullary thyroid cancer (MTC).
- MTC is a rare form of thyroid cancer
- In the United States, about 1200 cases are diagnosed each year
- 80% are sporadic
- 20% are familial
- MTC arises from parafollicular cells
- The tumors secrete calcitonin (a useful tumor marker)
- Survival at 10 years: 65%
- Survival at 10 years with distant metastasis: 40%
The TNM staging system is the standard classification system used to estimate survival in many cancers, including medullary thyroid cancer. This system allows physicians to define tumor characteristics necessary to plan treatment, estimate the patient’s prognosis, and provide a method for researchers to exchange patient information.
- T = tumor size
- N = presence or absence of metastatic disease
- M = presence or absence of distant metastatic disease
Besides being a static system, Dr. Ganly explained the TNM staging system does not take into account predictive factors important in MTC, such as the patient’s age or gender (male gender is associated with poorer outcomes). Furthermore, TNM does not consider biological factors such as RET mutation status, calcitonin levels, or carcinoembryonic antigen (CEA) blood levels. Other variables such as margin status, vascular invasion, mitotic rate, and differentiation—which can be important predictors of outcome, are not considered in the TNM staging system.
A nomogram is a statistical tool that can help predict outcomes in individual patients; not patient populations, as the TNM does. Nomograms are being utilized to predict outcomes in patients with breast cancer and prostate cancer—and are shown to be superior to the TNM staging system.
“So the aim of our study was to try to design a predictive nomogram for medullary thyroid cancer and for cancer-specific mortality using clinical variables such as TNM, using pathological variables, and also using biochemical variables such as calcitonin,” stated Dr. Ganly.
The study subjects were patients with medullary thyroid cancer from MSKCC that were identified from the period of 1986-2010. Patients with unresectable disease were excluded leaving 249 patients for analysis. The patients’ tumors and tumor characteristics were reported and variables predictive of cancer-specific mortality (CSM) were identified by univariable analyses. A multivariable competing risk model was then developed to predict 10-year cancer-specific mortality. Predictors included age, gender, serum calcitonin (pre- and post-op), CEA, RET mutation status, vascular invasion, margin status, and mitotic rate.
- 66% of the patient were >45 years of age
- 50% female
- 16% had a RET mutation
- 42% had T3 or T4 tumors
- >50% pathologically positive disease
- 10% distant metastases
- 38% vascular invasive
- Pre-op calcitonin level: 1305
- CEA levels: median pre-op = 40
- CEA levels: median post-op = 2.8
- 56 deaths (22%) at 87 months
“Our final model could only contain seven variables based on the number of events and the number of patients. And the most predictive variables for our nomogram were age, gender, pre- and post-op calcitonin, TNM, and vascular invasion,” stated Dr. Ganly. “For each variable, you determine the number of points for each variable … The points for each variable are added up to create a total number of points. And using the total number of points we can calculate the probability of 10 year cancer-specific mortality,” explained Dr. Ganly. The nomogram Dr. Ganly describes has limitations because it was developed from retrospective data.
January 27, 2015