CHICAGO — Thyroid-stimulating hormone receptor messenger RNA is a clinically useful blood test in the pre- and postoperative management of thyroid cancer, based on results from a prospective validation study in 1,095 consecutive patients.
If detectable preoperatively, when fine-needle aspiration suggests follicular neoplasm, thyroid-stimulating hormone receptor (TSHR) messenger RNA (mRNA) accurately predicts thyroid cancer and guides the extent of surgery, reported Dr. Mira Milas, principal investigator.
“It has a practical role to predict thyroid cancers where other modalities may fall short,” she said. In addition, high levels on the first postoperative day alert physicians to persistent disease, while its presence during long-term surveillance aids in identifying recurrence.
“We are encouraged by this performance and the fact that it is a convenient blood test to perform,” said Dr. Milas, director of the thyroid center at the Cleveland Clinic. “Our next steps are to search for multicenter trials and [Food and Drug Administration] approval for some of these indications.”
No new blood tests for differentiated thyroid cancer have been introduced into routine clinical practice since thyroglobulin. Thyroglobulin is used for cancer follow-up, but not initial diagnosis.
The TSHR mRNA assay was developed 9 years ago at the clinic's pathology department by study collaborator Manjula Gupta, Ph.D.
TSHR mRNA acts as a surrogate marker for circulating thyroid cancer cells, and in initial studies distinguished benign from malignant thyroid diseases.
A recent report suggests that TSHR mRNA is detectable even in thyroid microcarcinomas, and may characterize those with potentially more aggressive histology (Surgery 2009;146:1081–9).
The aim of the current analysis was to validate the clinical use of the marker 1 year after its introduction at the Cleveland Clinic, where the test is now used daily in all patients scheduled for thyroid surgery and those undergoing consultation for thyroid disease in the office.
From October 2008 through September 2009, TSHR mRNA was measured by quantitative real-time polymerase chain reaction from blood drawn in 403 patients undergoing thyroid surgery, postoperatively in 541 patients, and in 151 patients monitored for benign goiters.
Preoperative Use
Preoperative TSHR mRNA greater than 1 ng/mcg as a sole predictor of cancer had a positive predictive value (PPV) of 81% and specificity of 83% in 374 patients with surgically confirmed pathology, Dr. Milas said. Sensitivity was modest at 61%, as was negative predictive value at 64%. However, its PPV was 100% in patients with papillary thyroid cancers greater than 1 cm.
She observed that TSHR mRNA is particularly useful in detecting cancer in patients with follicular neoplasms on fine-needle aspiration. In 54 such patients, TSHR mRNA alone had a PPV of 96%, a specificity of 96%, and accuracy of 85%. The sensitivity of diagnosing cancer improved from 76% to 97% when the blood test was combined with ultrasound. Ultrasound features such as irregular margins, hypervascularity, indistinct borders, and microcalcification are suggestive of cancer in follicular neoplasms, but none are independently diagnostic. The highest risk of a false positive occurs with Hashimoto's disease, Dr. Milas acknowledged.
Postoperative Use
Use of the blood test in 69 paired samples with thyroid cancer revealed that elevated TSHR mRNA levels became undetectable in all patients on postoperative day 1, except in seven who had persistent or recurrent cancer within the year and unfavorable histologic features.
In contrast, all 40 patients with benign disease had undetectable TSHR mRNA levels on day 1 after thyroidectomy. “This suggests the potential use of TSHR mRNA as an early marker of adequate surgical clearance of disease or future recurrence,” Dr. Milas said in an interview.
Long-Term Surveillance
TSHR mRNA also showed merit during long-term follow-up of thyroid cancer, notably in two particularly challenging scenarios: when thyroglobulin antibodies were present, and when patients had residual thyroid tissue from original surgery and detectable thyroglobulin levels between 0.2 and 49 ng/mL that are difficult to interpret, she said.
At a median follow-up of about 2 years in 60 patients with elevated thyroglobulin antibodies, detectable TSHR mRNA was the only blood test to confirm cancer in four patients. Negative TSHR mRNA reassured absence of disease in 54 of the 56 remaining patients whose imaging was also negative.
All 59 patients with residual thyroid tissue because of partial thyroidectomy for cancer had undetectable TSHR mRNA levels and no radiologic evidence of cancer recurrence at follow-up, she said.
“This assay has come of age,” remarked Dr. Collin Weber, the invited discussant and chief of endocrine surgery at Emory University Hospital in Atlanta.
He questioned whether total thyroidectomy is performed at the Cleveland Clinic on all indeterminate lesions when the assay is positive and whether its accuracy is good enough to recommend observation of TSHR mRNA-negative follicular neoplasms greater than 2.5 cm in size.