BOSTON – An analysis of data from medullary thyroid cancer patients that partitioned the patients into groups with similar overall survival has spurred a rethink of the current American Joint Committee on Cancer (AJCC) staging system.
The results from researchers at Duke University, Durham, N.C., presented at the annual meeting of the Endocrine Society by Dr. Mohamed Abdelgadir Adam, are timely, as the AJCC has embarked on a reconsideration of the staging of cancers, including medullary thyroid cancer (MTC), as part revisions for the eighth edition of the staging system.
“The existing AJCC staging system for MTC appears to be less than optimal in discriminating the risk of mortality among disease stage groups,” said Dr. Adam, who discussed the findings in a video interview.
MTC, a neuroendocrine tumor that affects C cells of the thyroid, comprises 3%-5% of all cases of thyroid cancer and it can be a more aggressive disease than differentiated thyroid cancer. Yet the current AJCC MTC staging system has been extrapolated from differentiated thyroid cancer data.
“We sought to evaluate how well the current AJCC seventh edition stage groupings predict survival for patients with MTC, to suggest a possible staging revision to sharpen estimates of prognosis,” said Dr. Adam.
The researchers utilized the National Cancer Data Base, representing over 70% of incident cancer cases in the United States.
MTC patients who underwent thyroid surgery from 1998 to 2012 were identified. Patients with missing values for pathologic T, N, or M were excluded. The primary outcome in the 3,315 patients was survival.
The researchers used a form of decision-tree analysis called recursive partitioning. In general, recursive partitioning is able to classify a population by splitting subjects into subgroups, each of which is homogeneous based on the particular outcome. In this study, the subgroup allocations were based on T, N, and M stages, with the outcome being overall survival. Kaplan-Meier and adjusted survival analyses enabled survival differences among the four subgroups (groups I, II, III and IV) to be explored.
The four groups were distinct in terms of survival time and allowed more accurate risk stratification. In particular, groups I and II were markedly better distinguished from one another than is the case with the current staging system. Survival differences across the stages were more distinct with the newly created T, N, and M groupings, compared with the current AJCC staging system.
After adjustment, survival differences across TNM groups were more distinct with the newly created TNM groupings (compared to subgroup I, hazard ratio of 3.06 for subgroup II; HR, 6.79 for III; and HR, 17.03 for IV), compared with the current AJCC staging (compared to stage I, HR, 1.45 for stage II; HR, 2.17 for III; and HR, 5.33 for IV).
“The AJCC is reevaluating all staging schemas, including MTC. The current AJCC staging system could be improved with the newly identified TNM groupings suggested here for more accurate patient risk stratification and possibly treatment selection,” said Dr. Adam.
Dr. Adam had no disclosures.