PerspectiveIn the Journals Plus

Malignant thyroid nodules grow faster than benign nodules

Show Citation

October 18, 2017

Erik K. Alexander
Erik K. Alexander

Malignant thyroid nodules are more likely to grow at least 2 mm per year and increase in volume compared with benign thyroid nodules, according to findings published in The Journal of Clinical Endocrinology & Metabolism.

Erik K. Alexander, MD, chief of the thyroid section at Brigham and Women’s Hospital, co-director of the Endocrine Cancer Treatment Center and professor of medicine at Harvard Medical School, and colleagues evaluated data from adults with 126 malignant thyroid nodules (mean age, 48.6 years; 15.9% men) and 1,363 benign thyroid nodules (mean age, 52.2 years; 9.9% men) to compare the growth rates of the nodules and the potential clinical relevance.

Participants with malignant nodules were included if they had two or more ultrasound assessments at least 6 months apart before surgical resection; participants with benign nodules were included if they had two or more ultrasound assessments at least 1 year apart. Median time between the two ultrasound assessments was 20.9 months in the malignant group and 21.8 months in the benign group.

Growth rate was greater for malignant nodules compared with benign nodules; a growth rate of at least 2 mm per year was present in 26.2% of the malignant nodules vs. 11.7% of benign nodules (P < .0001). During follow-up, 88.3% of benign nodules were stable or smaller vs. 73.8% of malignant nodules (P < .001). A growth rate of more than 2 mm per year was independently associated with malignancy (P < .01).

Nodules that grew more than 2 mm to 4 mm per year had an RR for malignancy of 1.85 (P = .001), and this RR increased with each increment of growth. For example, those growing more than 8 mm per year had an RR of 5.05 (P < .01) compared with stable nodules.

A greater than 20% size increase in two or more dimensions was observed in more malignant nodules (25.4%) than benign nodules (14.2%; P < .001), during similar median follow-up intervals. Additionally, more malignant nodules increased more than 50% in volume compared with benign nodules (34.9% vs. 21.1%; P < .001).

“These data demonstrate that clinically relevant ( 1 cm) cancerous thyroid nodules grow more often, and grow faster, than clinically relevant ( 1 cm) benign thyroid nodules,” the researchers wrote. “Furthermore, aggressive thyroid cancers demonstrated the fastest growth over time. Conversely, nodules that are stable, and especially those decreasing in size, are much more likely to prove benign. Together, these data suggest that thyroid nodule growth (and especially rapid nodule growth) should be reintroduced as an important variable in the evaluation of follow-up of thyroid nodules.” – by Amber Cox

Disclosures: The authors report no relevant financial disclosures.

itj+ Infographic

itj+ Perspective

Carolyn Maxwell

As nearly half the population have thyroid nodules, and with mounting evidence that we are overdiagnosing nodules as well as thyroid cancers, tools that help the endocrinologist noninvasively discern malignancy from the much more common benign nodule are in high demand. Recent evidence and published guidelines have suggested that sonographic features of thyroid nodules may be more specific than nodule growth rate at predicting malignancy. This is largely based on studies reporting that growth among benign nodules is not uncommon.

This study brings focus back to the use of growth rate as an independent predictor of malignancy. It is considerably well powered, with analysis of nearly 1,500 nodules, but its main strength is the comparison of growth rates of malignant nodules with benign nodules, a feature lacking in many long-term growth studies, for obvious reasons. Prospective size data are not generally available for malignant nodules, as they are typically resected; however, a large database of nodules at the Brigham and Women’s Hospital allowed for collection of nodule-size data of malignant nodules over several measurements prior to surgical removal. Careful consideration was given to the possibility of selection bias for this pool of nodules, and the authors, in my mind, have satisfactorily addressed the low likelihood that a bias of this sort could be influencing the data.

The results show that nodule growth, defined as greater than 2 mm per year, was indeed an independent predictor of malignancy and that faster growth rates increased the risk. Furthermore, cancers with higher-risk subtype (eg, medullary thyroid cancer, tall cell variant of papillary) were even more likely to demonstrate growth over the follow-up period. As we are using more selective criteria to determine which nodules to biopsy, as well as the recent trend toward consideration of observing small known cancers, this study suggests that tracking nodule growth rate is a useful tool in determining appropriate care.

Carolyn Maxwell, MD, ECNU

Associate Fellowship Program Director
Assistant Professor of Medicine
Division of Endocrinology
Stony Brook Medicine

Disclosure: Maxwell reports no relevant financial disclosures.