A third of follicular thyroid lesions of undetermined significance were malignant

CHICAGO – At least in some institutions, about a third of follicular thyroid lesions of undetermined significance are malignant, reported researchers from the University of Wisconsin, Madison.

That rate is a higher proportion than the 5%-15% rate estimated by the Bethesda System, a national standard for reporting thyroid cytopathology, said Dr. Juan Carlos Jaume at the joint meeting of the International Congress of Endocrinology and the Endocrine Society.

"The message here is to be aware of your local institutional rates for cancer in FLUS, because following [the Bethesda System] may mislead you and your patient" when deciding on a course of action, be it surgery, repeat biopsy, or observation, he said. "Other institutions [should be encouraged to] do similar analyses to generate more accurate local guidelines for management of FLUS," noted Dr. Jaume, senior author of the study.

Of 1,420 nodules assessed over 2 years at the thyroid clinic at the University of Wisconsin, Madison, 134 (9.4%) were reported as follicular lesions of undetermined significance (FLUS) on fine-needle aspiration. Eighty patients opted for surgery; pathology revealed that 27 (34%) actually had differentiated thyroid cancer. Cancer also was found in four more patients, but at sites different from the original fine-needle aspiration. Most of the cancers (27) were papillary, 3 were follicular, and 1 was a Hurthle cell tumor, the investigators reported.

It’s unlikely the findings were due to selection bias, with patients who were more likely to have cancer opting for surgery. More than half of the patients chose surgery after discussing risks and benefits with their providers, not because of tumor progression. Almost all the others opted for surgery because of compression symptoms or because they had a nodule larger than 4 cm.

Even if there was a bias, "the highest expectation [with Bethesda] is 15%; our rate was 34%," a large difference, Dr. Jaume said. "As soon as we had the rate available, we conveyed the information" to providers so they could more accurately counsel patients. "I think eventually we will see an increase in the number of patients deciding on surgery."

The team performed the study because providers at the university had been relying on the Bethesda estimate to guide patients, but had a hunch that their local FLUS cancer rates were higher.

The majority of the 134 FLUS patients who opted against surgery chose ultrasound monitoring. Among the 22 who chose repeat fine-needle aspiration, half were rediagnosed with benign cytology, 5 were again diagnosed with FLUS, and most of the rest were lost to follow-up.

Among the 80 surgical patients, pathology was benign in 47 and parathyroid tissue was present in 1 biopsy. Records were unavailable for the final patient.

Some cytopathologists tend to call thyroid lesions FLUS more frequently than others; possibly, that predilection has something to do with the discordance in reported cancer rates, Dr. Jaume said.

Ultimately, the solution will be genetic analysis of fine-needle aspiration samples. There is a commercial product on the market, but "we are not using [it] in our institution because the negative predictive value is high, but the positive predictive value is low," he said.

The investigators had no relevant disclosures and had no outside funding for their work.

aotto@frontlinemedcom.com

*Correction, 8/25/2014: An earlier version of this article misspelled Dr. Jaume's name.

CHICAGO – At least in some institutions, about a third of follicular thyroid lesions of undetermined significance are malignant, reported researchers from the University of Wisconsin, Madison.

That rate is a higher proportion than the 5%-15% rate estimated by the Bethesda System, a national standard for reporting thyroid cytopathology, said Dr. Juan Carlos Jaume at the joint meeting of the International Congress of Endocrinology and the Endocrine Society.

"The message here is to be aware of your local institutional rates for cancer in FLUS, because following [the Bethesda System] may mislead you and your patient" when deciding on a course of action, be it surgery, repeat biopsy, or observation, he said. "Other institutions [should be encouraged to] do similar analyses to generate more accurate local guidelines for management of FLUS," noted Dr. Jaume, senior author of the study.

Of 1,420 nodules assessed over 2 years at the thyroid clinic at the University of Wisconsin, Madison, 134 (9.4%) were reported as follicular lesions of undetermined significance (FLUS) on fine-needle aspiration. Eighty patients opted for surgery; pathology revealed that 27 (34%) actually had differentiated thyroid cancer. Cancer also was found in four more patients, but at sites different from the original fine-needle aspiration. Most of the cancers (27) were papillary, 3 were follicular, and 1 was a Hurthle cell tumor, the investigators reported.

It’s unlikely the findings were due to selection bias, with patients who were more likely to have cancer opting for surgery. More than half of the patients chose surgery after discussing risks and benefits with their providers, not because of tumor progression. Almost all the others opted for surgery because of compression symptoms or because they had a nodule larger than 4 cm.

Even if there was a bias, "the highest expectation [with Bethesda] is 15%; our rate was 34%," a large difference, Dr. Jaume said. "As soon as we had the rate available, we conveyed the information" to providers so they could more accurately counsel patients. "I think eventually we will see an increase in the number of patients deciding on surgery."

The team performed the study because providers at the university had been relying on the Bethesda estimate to guide patients, but had a hunch that their local FLUS cancer rates were higher.

The majority of the 134 FLUS patients who opted against surgery chose ultrasound monitoring. Among the 22 who chose repeat fine-needle aspiration, half were rediagnosed with benign cytology, 5 were again diagnosed with FLUS, and most of the rest were lost to follow-up.

Among the 80 surgical patients, pathology was benign in 47 and parathyroid tissue was present in 1 biopsy. Records were unavailable for the final patient.

Some cytopathologists tend to call thyroid lesions FLUS more frequently than others; possibly, that predilection has something to do with the discordance in reported cancer rates, Dr. Jaume said.

Ultimately, the solution will be genetic analysis of fine-needle aspiration samples. There is a commercial product on the market, but "we are not using [it] in our institution because the negative predictive value is high, but the positive predictive value is low," he said.

The investigators had no relevant disclosures and had no outside funding for their work.

aotto@frontlinemedcom.com

*Correction, 8/25/2014: An earlier version of this article misspelled Dr. Jaume's name.

CHICAGO – At least in some institutions, about a third of follicular thyroid lesions of undetermined significance are malignant, reported researchers from the University of Wisconsin, Madison.

That rate is a higher proportion than the 5%-15% rate estimated by the Bethesda System, a national standard for reporting thyroid cytopathology, said Dr. Juan Carlos Jaume at the joint meeting of the International Congress of Endocrinology and the Endocrine Society.

"The message here is to be aware of your local institutional rates for cancer in FLUS, because following [the Bethesda System] may mislead you and your patient" when deciding on a course of action, be it surgery, repeat biopsy, or observation, he said. "Other institutions [should be encouraged to] do similar analyses to generate more accurate local guidelines for management of FLUS," noted Dr. Jaume, senior author of the study.

Of 1,420 nodules assessed over 2 years at the thyroid clinic at the University of Wisconsin, Madison, 134 (9.4%) were reported as follicular lesions of undetermined significance (FLUS) on fine-needle aspiration. Eighty patients opted for surgery; pathology revealed that 27 (34%) actually had differentiated thyroid cancer. Cancer also was found in four more patients, but at sites different from the original fine-needle aspiration. Most of the cancers (27) were papillary, 3 were follicular, and 1 was a Hurthle cell tumor, the investigators reported.

It’s unlikely the findings were due to selection bias, with patients who were more likely to have cancer opting for surgery. More than half of the patients chose surgery after discussing risks and benefits with their providers, not because of tumor progression. Almost all the others opted for surgery because of compression symptoms or because they had a nodule larger than 4 cm.

Even if there was a bias, "the highest expectation [with Bethesda] is 15%; our rate was 34%," a large difference, Dr. Jaume said. "As soon as we had the rate available, we conveyed the information" to providers so they could more accurately counsel patients. "I think eventually we will see an increase in the number of patients deciding on surgery."

The team performed the study because providers at the university had been relying on the Bethesda estimate to guide patients, but had a hunch that their local FLUS cancer rates were higher.

The majority of the 134 FLUS patients who opted against surgery chose ultrasound monitoring. Among the 22 who chose repeat fine-needle aspiration, half were rediagnosed with benign cytology, 5 were again diagnosed with FLUS, and most of the rest were lost to follow-up.

Among the 80 surgical patients, pathology was benign in 47 and parathyroid tissue was present in 1 biopsy. Records were unavailable for the final patient.

Some cytopathologists tend to call thyroid lesions FLUS more frequently than others; possibly, that predilection has something to do with the discordance in reported cancer rates, Dr. Jaume said.

Ultimately, the solution will be genetic analysis of fine-needle aspiration samples. There is a commercial product on the market, but "we are not using [it] in our institution because the negative predictive value is high, but the positive predictive value is low," he said.

The investigators had no relevant disclosures and had no outside funding for their work.

aotto@frontlinemedcom.com

*Correction, 8/25/2014: An earlier version of this article misspelled Dr. Jaume's name.