Outcome After Treatment of High-Risk Papillary and Non-Hurthle-Cell Follicular Thyroid Carcinoma

Taylor T, Maxon HR, et al.
Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 1, 2001

Reviewers: John Han-Chih Chang, MD
Source: Annals of Internal Medicine 1998; Volume 129 (Number 8): pages 622-627

OncoLink Journal Scans


Differentiated thyroid cancers are the most common type, making up over 90% of the cases. Papillary and follicular are the major subtypes under the differentiated heading. This prospective registration trial attempted to evaluate the efficacy of the different treatment modalities [surgery, radioactive iodine (RI) and external beam radiation therapy (EBRT)] for differentiated thyroid malignancies.

Materials and Methods

This is a prospective nonrandomized registration trial as acollaborative effort of the National Thyroid Cancer TreatmentCooperative Study Registry from 14 institutions. A total of 1607thyroid cancer patients were registered. Patients deemed to be at highrisk were collected to be part of the data set. The criteria forpapillary thyroid cancer was age 45 years or older and mass greaterthan 4 cm or extraglandular tumor extension or lymph node involvementOR distant metastases at any age. For follicular thyroid cancer, thehigh risk criteria was age 45 years or greater with tumor greater than1 cm or multifocality or capsular invasion or regional lymph nodemetastases or extraglandular extension OR any age with distantmetastases or poorly differentiated disease. Based on the abovecriteria, 303 patients with papillary and 82 patients with follicularcarcinoma met the high risk profile.


Mean follow up was 3 years. Mean age of entry was 57 years old. Most of the entrants into the study were women. Women with papillarycancer had a lower overall mortality (death rate) than men, but cancerspecific mortality (death from thyroid cancer) was not statisticallydifferent. Historically, older age had an association with increasedmortality from thyroid cancer, but none was noted in the studiedpopulation. This may be a bias introduced by the criteria used toselect the patients.

The initial extent of thyroid surgery (total thyroidectomy versusnear-total) did improve overall mortality for papillary thyroid cancer,but did not affect progression of disease or disease free survival. There was no effect on follicular thyroid cancers, either positive ornegative. The complication rate of surgery was approximately 14% withmost of the toxicity being hypoparathyroidism (5.5%) and vocal cordpalsy (7%).

RI was utilized in 85% of the patients postoperatively. Mostpatients received more than 75 mCi of activity (comparable to dose ofradiation). In papillary and follicular thyroid cancer, treatment withRI was of benefit in cancer specific survival and disease progression. Only in the follicular thyroid cases did the RI treatment improveoverall mortality.

Only 18% received EBRT for their thyroid cancer. The average dosewas 46 Gy at 2.5 Gy fractions. Approximately half had lymph nodemetastases at the time of surgery (about the same fraction of thenon-EBRT treated patients were lymph node positive). Those whoreceived EBRT did have gross residual disease more often (20%) thanthose who did not receive EBRT (9%). Gross extrathyroid invasion wasseen more often in the EBRT treated population (88%) versus those notreceiving EBRT (65%). EBRT treated patients were older (mean age 61years versus 55). Overall mortality was increased in the papillary andfollicular thyroid cancer patients receiving EBRT. Increased cancerspecific mortality, likelihood of progression and poorer disease freesurvival were seen in the follicular cancer population treated withEBRT.


The authors conclude that surgery should consist of total ornear total thyroidectomy since it is associated with reduced improvesoverall mortality in papillary disease. They also suggest that RIshould always be utilized, especially in follicular disease. Theauthors make a blanket statement stating that EBRT is associated with apoorer outcome.

This prospective multi-institutional registration trial is just athorough retrospective review. This in actuality is not much morepowerful a study than a retrospective review. There was norandomization and treatment was not standardized. This becomes anissue when one tries to make conclusions on what should be standard ofcare in the treatment of thyroid cancer. I would agree surgery isnecessary and that it needs to be as aggressive as possible to removethe iodine sink that would decrease the effectiveness of RI. I alsoagree that RI is an effective adjunct to treatment in these patients. But for the amateur reader, this article may sway a clinician away fromusing EBRT in patients with differentiated thyroid cancer. I agreeEBRT is not indicated in all individuals with differentiated thyroidcancer. But, EBRT is ASSOCIATED with a worse outcome, because of thepoor prognostic patients it has to treat. EBRT does not make theoutcome worse as a conclusion to which this article seems to allude. As noted above the patients treated with EBRT had a higher rate ofgross residual disease, gross extrathyroidal invasion and older age. These patients require more treatment and to believe otherwise would bea disservice to the patient. Other minor criticisms are that follow upis too short and patient numbers are too small to make too many widesweeping conclusions about each of these treatment modalities.

The true answer on how best to treat differentiated thyroid cancersshould be further pursued since its incidence is steadily growing. Aprospective RANDOMIZED trial needs to be performed in order to fullyevaluate the benefit of each modality, especially EBRT. Only then willwe have the real answer. Glorified retrospective reviews are onlyprovide some small hindsight into the issues at hand.