DOI:10.4158/EP151006.OR © 2015 AACE.

ENDOCRINE PRACTICE Rapid Electronic Article in Press Rapid Electronic Articles in Press are preprinted manuscripts that have been reviewed and accepted for publication, but have yet to be edited, typeset and finalized. This version of the manuscript will be replaced with the final, published version after it has been published in the print edition of the journal. The final, published version may differ from this proof. DOI:10.4158/EP151006.OR © 2015 AACE.

Original Article EP151006.OR

LONG-TERM NON-OPERATIVE RATE OF THYROID NODULES WITH BENIGN RESULTS ON THE

AFIRMA GENE EXPRESSION CLASSIFIER

Jennifer A. Sipos, MD1; Thomas C. Blevins, MD2; Heidi Chamberlain Shea, MD, FAAP, FACE3;

Daniel S. Duick, MD, FACE, FACP4, Shamsher K. Lakhian, MD, FRCP(C), FACE3; Brian E.

Michael, MD5; Michael J. Thomas, MD, PhD6; Julie Ann Sosa, MD, MA, FACS7

Running title: Surgical Rate of GEC Benign Nodules

From: 1The Ohio State University, Columbus, OH; 2Texas Diabetes and Endocrinology, Austin, TX; 3Endocrine Associates of Dallas, , Dallas, TX; 4Endocrinology Associates, P.C., Scottsdale, AZ; 5Wellspan Health, Gettysburg, PA; 6Carolina Endocrine, P.A., Raleigh, NC; 7Section of Endocrine Surgery, Duke University School of Medicine, Durham, NC.

Correspondence address: Jennifer A. Sipos, MD,

Associate Professor of Medicine,

The Ohio State University, Columbus, OH.

Email: Jennifer.Sipos@osumc.edu

DOI:10.4158/EP151006.OR © 2015 AACE.

ABSTRACT

Objective The primary objective was to assess the operative rate in patients with a benign

result from the Afirma Gene Expression Classifier (GEC) during long-term follow-up at non-

academic medical facilities. The secondary endpoint of this study was the treating physician’s

opinion regarding the safety of GEC use compared to the hypothetical situation of providing

thyroid nodule management without the GEC.

Methods This is a retrospective study of non-academic medical practices utilizing the GEC.

Those clinicians utilizing the GEC testing who had three or more ‘benign’ results during the data

collection period (Sept 2010 through June 2014) were invited to participate. Operative status

and patient demographics were documented for patients with GEC testing at least 36 months

(±3 months) prior to the date of data collection. A survey also was administered to the treating

physicians to assess their perceived safety of using the GEC in patient care.

Results During 36 months (+/- 3 months) of follow up, 17 of 98 patients (17.3%) with a

‘benign’ GEC result underwent surgery. Within the first two years after a ‘benign’ GEC, 88% of

surgeries were performed. Regarding safety of the GEC, the treating physicians reported that

patient safety was improved by using the GEC compared to not using the GEC in 78 of 91 (86%)

cases.

Conclusion It appears that a ‘benign’ result on the GEC is associated with a reduction in the

DOI:10.4158/EP151006.OR © 2015 AACE.

rate of thyroid surgeries compared to published data when patients are followed for 36

monthsafter testing. A non-operative approach to follow-up was felt to be a safe alternative to

diagnostic surgery by the majority of responsible physicians in the study.

Abbreviations: GEC = Gene Expression Classifier; FNA = fine needle aspiration; AUS = Atypia of undetermined significance; FLUS = Follicular lesion of undetermined significance; FN = follicular neoplasm; SFM = Suspicious for malignancy.

Introduction

Thyroid nodules are a common clinical entity, palpated in up to 5% of patients and identified by

ultrasound in nearly 60% of patients over the age of 70 years in the United States [1]. Since

thyroid cancer is identified in fewer than 10% of fine needle aspirations (FNA) [2, 3], an efficient

evaluation of patients with nodules is of paramount importance to avoid unnecessary testing

and, potentially, surgery, along with their attendant morbidity and cost. FNA allows the clinician

to stratify the malignancy risk of a nodule safely in the ambulatory setting at a

relatively low cost and with a high sensitivity. The creation of the Bethesda classification

system for standardization of reporting cytologic diagnoses allows for easy communication

between the cytopathologist and the clinician regarding the risk of malignancy and aids in

planning for surveillance or additional testing, as appropriate [4]. However, up to 30% of

thyroid FNAs have indeterminate cytology findings, as defined by the Bethesda criteria

subtypes of atypia of uncertain significance (AUS), follicular lesion of undetermined significance

DOI:10.4158/EP151006.OR © 2015 AACE.

(FLUS), follicular neoplasms (FN), or suspicious for malignancy (SFM) [4]. The rate of malignancy

reported in these lesions is highly variable, ranging from 5-45% for AUS/FLUS, 15-

50% for FN, and 50-85% for SFM [5]. Without additional tools for risk stratification, the clinician

is left to choose between potential undertreatment with observation alone versus over-

treatment with surgery.

DOI:10.4158/EP151006.OR © 2015 AACE.

The Afirma gene expression classifier (GEC) (Veracyte, Inc.) is a molecular test collected during

FNA that allows for the stratification of cytologically indeterminate nodules into one of two

messenger RNA expression molecular signature categories: ‘benign’ or ‘suspicious’ [6]. Those

lesions with a ‘suspicious’ result on the GEC have a malignancy rate of 17-52% [7-11]. In

contrast, those indeterminate lesions with a ‘benign’ GEC result have a likelihood of malignancy

comparable to that of a nodule with benign cytology on FNA and may be observed with serial

ultrasonographic examination [7].

Although the introduction of thyroid FNA dramatically reduced the number of diagnostic

surgeries, up to 16% of cytologically benign nodules are still surgically removed [5]. Common

reasons for surgery among patients with cytologically benign nodules includes compressive

symptoms, nodule size, enlarging nodule, concerns regarding cosmesis, suspicious ultrasound

characteristics, family history of thyroid cancer, history of external irradiation to the neck, new

patient anxiety, or physician preference. It would be expected that surgery among GEC

benign nodules may be reduced, but not completely eliminated. Short-term follow up after

the introduction of the GEC revealed that surgery in cytologically indeterminate nodules was

reduced from a historical rate of 74% to between 2.8-25% when the GEC result suggested a

benign lesion [8-14]. Overall 58 of these 715 (8.1%) nodules with a benign Afirma GEC result

went to surgery [8-14]. This lower rate of surgery was not significantly different from the

historically reported rate of surgery for cytologically benign nodules [12]. It is unclear if the

non-operative rate is sustained during long-term follow-up of GEC ‘benign’ nodules.

DOI:10.4158/EP151006.OR © 2015 AACE.

Additionally, the application of this testing outside of academic medical centers is unclear.

Since community physicians account for a greater proportion of both thyroid FNAs and Afirma

GEC use, and as they are less likely than their academic counterparts to publish their clinical

experience, we sought to examine the clinical utility of the Afirma GEC in the non-academic

setting. The goal of this study was to examine the rate of surgical intervention for patients with

a ‘benign’ result on the GEC test during long-term follow-up at non-academic medical practices.

The secondary endpoint of this study was to determine the treating physician’s opinion

regarding the safety of GEC use compared to the hypothetical situation of providing care

without the GEC.

Methods:

Study design

This is a retrospective, multicenter study designed by the principal investigators (JAS and JAS)

and the sponsor, Veracyte, Inc. Data collection and statistical analysis were provided by the

sponsor. Prior to initiating the study, the principal investigators (PIs) assumed responsibility for

the decision to publish. Throughout the study, the PIs maintained study oversight, and had

access to all collected data. The PIs wrote the manuscript and were not compensated in any

way for their roles in the study. The study protocol was approved by a central investigational

review board and was deemed to be of minimal patient risk. Waiver of informed consent was

granted.

DOI:10.4158/EP151006.OR © 2015 AACE.

The GEC is indicated in patients age ≥21 years with one or more sonographically-confirmed

thyroid nodule(s) ≥1 cm in diameter with indeterminate cytopathology (AUS/FLUS, FN, or SFN).

The FNA sample for GEC analysis is collected in a proprietary nucleic acid preservative and

shipped with frozen “cold bricks” at <25° C to Veracyte.

Physicians in practices that were utilizing the GEC test and which had received benign GEC

results on three or more occasions during the data collection period (September 2010 through

June 2014) were invited to participate. A minimum of three patients per endocrinology

practice with a benign GEC result were required for participation in the study to minimize the

risk of identifying individual patients based on the case report form data that were submitted.

Participating physicians requested copies of all benign GEC test results from the Veracyte

clinical laboratory for their patients. These reports aided the identification of patients for

inclusion and completion of the case report form data. No unique patient identifying

information was provided to the clinical study group.

Inclusion criteria

The following criteria were required for inclusion of patient data in the study. The index thyroid

FNA had to have been performed 3 or more years prior to the date of data collection. Patient

age at the time of FNA had to be 18 years or older. FNA was to have been performed on a

thyroid nodule that was ≥1cm in diameter. The thyroid nodule FNA had to have undergone

GEC testing with concomitant cytopathology examination performed by a central cytopathology

DOI:10.4158/EP151006.OR © 2015 AACE.

group, resulting in an indeterminate cytopathology interpretation (Bethesda III or IV) and a

‘benign’ GEC interpretation. Operative status was to have been documented in the medical

chart within the 36 months (±3 months) or longer post initial GEC testing.

Exclusion criteria Patients were excluded from enrollment if they had other nodule(s) in the thyroid with a

cytologic diagnosis of Bethesda category V or VI or with a GEC ‘suspicious’ result. Patients also

were excluded if their GEC test result was ‘no result’. Finally, patients were excluded if surgery

was already likely or planned, regardless of the FNA result; these might include patients with

compressive goiters or where patient preference already had guided management toward

surgery.

Data collection

After verification of patient eligibility, the individual practice sites recorded information about

patient demographics, thyroid nodule size and location, and whether thyroid surgery was

performed less than 6 months or more than 6 months after the index FNA. If surgery was

performed, the date of surgery, the surgery type (partial or total/near-total thyroidectomy),

and the reason for surgery were reported. The case report data form was pre-populated with

likely reasons for proceeding to surgery, including: large nodule, rapid nodule growth since

index FNA, new local compressive symptoms clearly associated with the nodule, suspicious

ultrasound characteristics, indeterminate cytology, family history of thyroid cancer, multiple

endocrine neoplasia syndrome, new suspicious lymph node, subsequent repeat FNA with a

DOI:10.4158/EP151006.OR © 2015 AACE.

suspicious for malignancy or malignant cytology, subsequent repeat FNA with GEC ‘suspicious’

result, newly described patient anxiety since performance of the index FNA, history of external

beam irradiation to the neck, or new concern since index FNA about cosmesis related to the

nodule. The treating physician was surveyed regarding patient safety by using the GEC

compared to the hypothetical scenario of not using the GEC with one of three questions: was

safety improved by using the GEC, diminished using the GEC, or it made no difference by using

the GEC? The definition of patient safety was intentionally vague for interpretation by the

provider.

Statistical Analysis

Kaplan-Meier analysis was performed to illustrate the outcome of surgical rate over time. To

overcome the potential for over-estimation of the surgical rate due to lack of follow up, two

estimates of the Kaplan-Meier analysis were performed. With the stringent censoring analysis,

subjects who were lost to follow-up or had surgery during a given time period were counted

among those who survived through the end of that time period; they were then deleted from

the number at risk for the next time period. In contrast, the loose censoring analysis included

patients as un-operated through 36 months when they did not have documentation of follow-

up but whose physician affirmatively responded “to the best of my knowledge, the patient

remains under my care and remains un-operated”. This method allowed patients to be included

despite not having returned for follow-up given potentially long intervals between prescribed

follow-up visits, all while remaining under a physician’s care.

DOI:10.4158/EP151006.OR © 2015 AACE.

To examine whether patient age, gender, or nodule size was a significant risk factor associated

with the surgery rate, a Cox proportional hazards model was used. Univariate analysis was

applied to compare the Surgery versus No Surgery groups; p values were two-sided, and

significance was defined as a value <0.05. All analyses were performed using R software

version R 3.1.0.

Results

A search of the Veracyte records of ‘benign’ GEC reports revealed 21 eligible practices; 17 were

IRB-approved. One IRB-approved site reported that all eligible patients were lost to follow up,

and this site was excluded. The four practices that did not participate indicated a lack of

support staff as the reason for inability to participate in the study. These practices had 29

potential patients with a mean nodule size of 2.51 cm recorded on the sample requisition

versus a mean size of 2.47 cm among the patients from the 17 IRB-approved sites, suggesting

no difference in nodule size between participating and non-participating practices. Similarly,

the four practices that did not participate were from a total of 3 states, and these states were

represented in the study by other practices. In total, 16 sites participated; 16 endocrinologists

and one radiologist submitted data on 98 patients (see Table 1). The median follow-up time

was 26 months (range 0-44 months). Thyroidectomy was performed in 17 patients (17.3%

operative rate).

Kaplan-Meier analysis demonstrated that the majority of surgeries (10 of 17, or 58.8%)

occurred within the first year after obtaining the GEC ‘benign’ result. Within two years of

DOI:10.4158/EP151006.OR © 2015 AACE.

receiving a ‘benign’ GEC result, 88% of all surgeries were performed. The rate appeared to level

off after the first year of follow-up (Figures 1a and 1b). In the second and third years after a

‘benign’ GEC result was obtained, an additional 5 and 2 patients underwent surgery,

respectively.

The most common indications for surgery were “rapid nodule growth” and “large nodule”

(Table 2). The development of new onset symptoms of compression was provided as the

rationale for the removal of two benign nodules. Two nodules had ultrasonographically

suspicious features that prompted their excision. Patient anxiety, substernal goiter extension,

and second opinion by a surgeon each were listed once as reasons for the removal of a ‘benign’

GEC nodule. These operations all occurred within 7.5 months of the index FNA.

To analyze whether patient age, gender, or thyroid nodule size were factors associated with

undergoing surgery subsequent to a ‘benign’ GEC result, Cox proportional hazards modeling

was used to compare the Surgery and No Surgery management groups. No significant

differences were seen between the groups with regard to patient age (p = 0.11 with stringent

and loose censoring) or gender (p = 0.87 with stringent censoring and p = 0.94 with loose

censoring). Larger nodule size was associated with the Surgery group compared to the No

Surgery group (p = 0.00020 with stringent censoring, 0.00022 with loose censoring). The mean

index nodule size among the operated and unoperated patients was 3.24 cm (2-5.9 cm) and

2.28 cm (1-7 cm), respectively (p = 0.00006).

DOI:10.4158/EP151006.OR © 2015 AACE.

When physicians were surveyed about the utility of the GEC testing, 78 of 91 (86%) physicians

reported that patient safety was improved by using the GEC test compared to not using it. In

comparison, 11 out of 91 physicians (12%) reported that using the GEC test had no impact on

perceived patient safety. Two physicians indicated that patient safety was diminished using the

GEC.

Discussion

The arrival of the era of molecular testing brings with it questions and opportunities regarding

the long-term management and follow up of patients with molecularly tested thyroid nodules.

The objective of molecular testing with the GEC test is to identify benign nodules among those

with indeterminate cytology, thereby circumventing a diagnostic thyroidectomy. With a

reported surgical rate of 74% among cytologically indeterminate nodules [12], GEC testing

offers an opportunity to potentially avoid a diagnostic operation when in fact the indeterminate

nodule is likely benign. Several studies, mostly from academic medical centers and with

relatively short durations of follow-up, have demonstrated that use of the GEC is associated

with a reduction in diagnostic thyroidectomies by nearly 50% [8-14]. We sought to determine if

avoidance of surgery was facilitated with a benign GEC result in the community practice setting,

and to investigate if the decision to avoid surgery was durable with longer-term follow up.

Our study revealed that the majority of patients with long-term follow up after a benign GEC

remained in surveillance without an operation. The observed surgical rate of 17.3% is similar to

DOI:10.4158/EP151006.OR © 2015 AACE.

that seen in other studies of its kind, including multicenter short-term follow up studies by

Duick et al [12] and Alexander et al [8]. In the trial by Duick et al [12] the operative rate among

360 nodules with a ‘benign’ GEC was 7.6%. This surgical rate was not statistically different than

the rate of surgery on cytologically benign nodules at 9% [12]. Similarly, the independent and

multicenter analysis from five academic institutions reported 174 patients with a ‘benign’ GEC

result, of which 71 that were followed for a median of 9 months. Overall, 11 (6.3%) GEC benign

patients underwent surgery. Indications for surgery were similar to those seen in the present

study, including physician recommendation (4 nodules), compressive symptoms (4 nodules),

and cosmetic concerns (3 nodules) [8]. The present study is unique in its long-term follow up,

with a median of 26 months (range 0-44 months). The decreasing rate of surgery over the first

36 months of follow up, combined with the observation that 88% of all surgeries occurred

within the first two years, suggests that studies with longer follow-up duration are unlikely to

give dramatically different rates of surgery among GEC benign thyroid nodules.

Among patients undergoing surgery in spite of a ‘benign’ GEC result, new compressive

symptoms and/or rapid nodule growth impacted the decision to operate nearly half the time.

Large nodule size was listed as the primary factor in the decision to proceed to surgery in 4 of

17 patients. Indeed, a key finding in this study was that nodules surgically removed were 1 cm

larger than unoperated nodules. Other authors have reported a similar experience in GEC

“benign” nodules, with the majority of indications for surgery centered on nodule size or

compressive symptoms [8, 12].

DOI:10.4158/EP151006.OR © 2015 AACE.

Given that the GEC sensitivity and NPV are imperfect, the treating physician must interpret the

results in light of the clinical situation. Prevalence of malignancy per Bethesda category within

an institution is relevant to the population, but institutional averages are much less relevant to

the individual patient who either has cancer or does not. Personalized medicine requires

shifting from prevalence of disease in the population to the risk of disease in the unique patient.

Individual clinical factors such as the nodule ultrasound characteristics help re- characterize

the individual patient’s risk. In this study, two patients were referred to surgery for

suspicious ultrasound findings. Decision-making based on the clinical features of the individual

patient can therefore help physicians practice personalized medicine and interpret results of

diagnostic testing within that framework.

The duration of time that a cytologically benign nodule should be followed is unclear. The rate

of identification of a false negative benign cytology is very low beyond three to four years of

follow up [15-17]. Further, the outcomes of malignancies that are grossly confined to the

thyroid are excellent, even with significant treatment delays [17-19]. A similar, low rate of

clinically significant false-negative malignancies is expected after three years of clinical

observation among nodules with indeterminate cytology and ‘benign’ GEC results[20]. Thus, if

the decision not to operate on nodules with GEC ‘benign’ results is durable through three years

of observation, broad conclusions regarding safety and durability of surveillance are unlikely to

significantly change in studies of longer duration.

This study has some important limitations, including its retrospective design. Additionally,

DOI:10.4158/EP151006.OR © 2015 AACE.

inclusion of patients was based on the treating physician’s willingness to participate, presenting

a potential source of bias. The loss of patients to follow up posed a challenge in ascertaining the

rate of surgery. We attempted to address this issue by creating two Kaplan-Meier curves, one

of which included those patients who had not had recent contact with their treating physician in

the group that had not had surgery (loose censoring), and another curve that removed those

lost patients from the analysis at the time of their last follow up (stringent censoring). These

two calculations were intended to provide the range at which surgery is performed in this long-

term follow up. Though not necessarily a limitation, it is important to note that this study was

not designed or undertaken as a validation trial. For this reason, the results of the surgical

histology were not collected for analysis.

An area of future research includes the natural history of Afirma ‘benign’ follicular neoplasms

and whether they might convert to a ‘suspicious’ GEC profile if left in situ for a period of time.

Indeed, there is a scarcity of data regarding nodules with repeat Afirma testing. Some believe

that a follicular adenoma could transform to a carcinoma if left alone; the natural history of

such lesions is unknown since historically, almost all have been surgically removed [21-24]. The

question remains whether a repeat GEC should be performed at some interval in the future,

depending on the sonographic features and potentially growth of the nodule.

DOI:10.4158/EP151006.OR © 2015 AACE.

Conclusion

This study demonstrates that a ‘benign’ GEC result is associated with a low operative rate. The

majority of surgeries occur in the first 12 months after a ‘benign’ GEC test; very few surgeries

were performed beyond two years of follow up. The majority of clinicians using the GEC felt

that it improved patient safety compared to not using it. Therefore, it appears that GEC testing

is associated with a durable reduction in the rate of surgery that clinicians believe is safe when

patients are followed for 36 months after a ‘benign’ GEC result.

References

1. Mazzaferri EL. Management of a solitary thyroid nodule. N Engl J Med. 1993; 328: 553-9.

2. Yang J, Schnadig V, Logrono R, Wasserman PG. Fine-needle aspiration of thyroid nodules: a

study of 4703 patients with histologic and clinical correlations. Cancer. 2007;111:306-15.

3. Yassa L, Cibas ES, Benson CB, et al. Long-term assessment of a multidisciplinary

approach to thyroid nodule diagnostic evaluation. Cancer. 2007;111:508-16.

4. Baloch ZW, Cibas ES, Clark DP, et al. The National Cancer Institute Thyroid fine needle

aspiration state of the science conference: a summation. Cytojournal. 2008 ;7;5:6.

5. Wang CC, Friedman L, Kennedy GC, et al., A large multicenter correlation study of thyroid

DOI:10.4158/EP151006.OR © 2015 AACE.

nodule cytopathology and histopathology. Thyroid. 2011;21:243-51.

6. Chudova D, Wilde JI, Wang ET et al. Molecular classification of thyroid nodules using high-

dimensionality genomic data. J Clin Endocrinol Metab.2010;95: 5296-304.

7. Alexander EK, Kennedy GC, Baloch ZW et al., Preoperative diagnosis of benign thyroid

nodules with indeterminate cytology. N Engl J Med. 2012; 367:705-15.

8. Alexander EK, Schorr M, Klopper J, et al. Multicenter clinical experience with the Afirma gene

expression classifier. J Clin Endocrinol Metab. 2014; 99:119-25.

9. Harrell RM, Bimston DN. Surgical utility of Afirma: effects of high cancer prevalence and

oncocytic cell types in patients with indeterminate thyroid cytology. Endocr Pract. 2014; 20:364-

9.

10. Lastra RR, Pramick MR, Crammer CJ, LiVolsi VA, Baloch ZW. Implications of a suspicious afirma

test result in thyroid fine-needle aspiration cytology: an institutional experience. Cancer

Cytopathol. 2014; 122:737-44.

11. McIver B, Castro MR, Morris JC, et al. An independent study of a gene expression classifier

(Afirma) in the evaluation of cytologically indeterminate thyroid nodules. J Clin Endocrinol

Metab. 2014;99:4069-77.

12. Duick DS, Klopper JP, Diggans JC, et al. The impact of benign gene expression classifier test

results on the endocrinologist-patient decision to operate on patients with thyroid nodules with

indeterminate fine-needle aspiration cytopathology. Thyroid. 2012. 22: 996-1001.

13. Marti JL, Avadhani V, Donatelli LA, et al., Wide Inter-institutional Variation in Performance of a

Molecular Classifier for Indeterminate Thyroid Nodules. Ann Surg Oncol. 2015;22:3996-4001.

14. Sullivan PS, Hirschowitz SL, Fung PC, Apple SK. The impact of atypia/follicular lesion of

DOI:10.4158/EP151006.OR © 2015 AACE.

undetermined significance and repeat fine-needle aspiration: 5 years before and after

implementation of the Bethesda System. Cancer Cytopathol. 2014; 122:866-72.

15. Durante C, Costante G, Lucisano G, et al. The natural history of benign thyroid nodules. JAMA.,

2015;313:926-35.

16. Lee S, Skelton TS, Zheng F, et al. The biopsy-proven benign thyroid nodule: is long-term follow-

up necessary? J Am Coll Surg.2013;217:81-8; discussion 88-9.

17. Nou E, Kwong N, Alexander LK et al. Determination of the optimal time interval for repeat

evaluation after a benign thyroid nodule aspiration. J Clin Endocrinol Metab. 2014; 99:510-6.

18. Amit M, Rudnicki Y, Binenbaum Y, et al. Defining the outcome of patients with delayed

diagnosis of differentiated thyroid cancer. Laryngoscope. 2014;124:2837-40.

19. Davies L, Welch HG. Thyroid cancer survival in the United States: observational data from

1973 to 2005. Arch Otolaryngol Head Neck Surg. 2010;136:440-4.

20. Michael B. Long Term Clinical and Imaging Follow-up of an Office-Based Gene Expression

Classifier Used to Manage Thyroid Nodules, in 84th Annual Meeting of the American

Thyroid Association. 2014: Coronado, CA. p. A-106.

21. Nikiforov, YE, Yip L, Nikiforova MN. New strategies in diagnosing cancer in thyroid

nodules: impact of molecular markers. Clin Cancer Res. 2013. 19:2283-8.

22. Fagin JA, Minireview: branded from the start-distinct oncogenic initiating events may

determine tumor fate in the thyroid. Mol Endocrinol. 2002; 16:903-11.

23. Zhu Z, Gandhi M, Nikiforova MN, Fischer AH, Nikiforov YE. Molecular profile and clinical-

pathologic features of the follicular variant of papillary thyroid carcinoma. An unusually high

prevalence of ras mutations. Am J Clin Pathol. 2003; 120: 71-7.

DOI:10.4158/EP151006.OR © 2015 AACE.

24. Burns JS, Blaydes JP, Wright PA, et al. Stepwise transformation of primary thyroid epithelial

cells by a mutant Ha-ras oncogene: an in vitro model of tumor progression. Mol Carcinog.

1992; 6:129-39.

DOI:10.4158/EP151006.OR © 2015 AACE.

Table 1 Site number State Number of patients enrolled 1 AZ 20 2 TX 6 3 NC 10 4 FL 6 5 TX 6 6 AK 3 7 MD 5 8 CO 5 9 TX 3 10 WA 4 11 FL 4 12 PA 3 13 FL 2 14 WA 3 15 NC 4 16 TX 11

DOI:10.4158/EP151006.OR © 2015 AACE.

Table 2

Subject Nodule size (cm) Time (months) Reason for surgery

26-01 2.3 5.9 New local pressure symptoms

12-06 3.1 1.2 Large nodule, New local pressure symptoms

32-11 2 23.2 Rapid nodule growth

32-09 2.5 16 Rapid nodule growth

13-03 4.2 8 Rapid nodule growth

25-01 2.2 2.4 Rapid nodule growth

12-02 2.19 30.1 Rapid nodule growth

17-03 3.4 14 Rapid nodule growth

28-01 4.13** 23 Large nodule, Ultrasound suspicious

11-07 2.15 3.3 Ultrasound suspicious

31-03 2.8 9.2 Large nodule

11-08 3.82 25.9 Large nodule

DOI:10.4158/EP151006.OR © 2015 AACE.

24-03 4 12.4 Large nodule

17-02 5.9 6.6 Large nodule

DOI:10.4158/EP151006.OR © 2015 AACE.

13-06 2.4 7.5 Other: Sub-sternal goiter extension*

11-17

5

2.2

Other: Second opinion by a surgeon*

30-04

3

1.2

New patient anxiety*

*The decision to operate was not made by the initially treating physician for 3 patients.

Prevalence of surgery according to the treating physician’s decision was 14/98 (14%).

** Patient had second GEC B nodule with suspicious ultrasound characteristics but the decision

to operate was based off the 4.13 cm nodule

DOI:10.4158/EP151006.OR © 2015 AACE.

Figure 1a. Kaplan Meier Curve Loose Censoring. Includes patients as un-operated through 36 months when they did not have documentation of follow-up but whose Physician affirmatively responded “To the best of my knowledge, the patient remains under my care and remains un-operated.”

DOI:10.4158/EP151006.OR © 2015 AACE.

DOI:10.4158/EP151006.OR © 2015 AACE.

Figure 1b. Kaplan Meier Curve Stringent Censoring. Subjects who were lost to follow up or had surgery during a given time period were counted among those who survived through the end of that time period; they were then deleted from the number at risk for the next time period.

DOI:10.4158/EP151006.OR © 2015 AACE.