Decision Making for

D i a g n o s i s an d
Management
Algorithms from Experts for Molecular Testing

Jeffrey Bumpous, MDa, Miranda D. Celestre, MDa,

Edmund Pribitkin, MDb,*, Brendan C. Stack Jr, MDc

KEYWORDS
! Thyroid nodule ! Thyroid cancer ! Ultrasonography
! Fine-needle aspiration cytology ! Biomarkers ! Molecular classifier

KEY POINTS
! Assessment of risk of thyroid nodules requires understanding of clinical, demographic,
imaging, cytopathologic, and now biomarker profiles; none of these factors alone repre-
sents a sufficient decision-making factor.
! Ultrasonography represents an accurate and cost-effective imaging modality for evalu-
ating the thyroid, cervical lymphatics, and postoperative thyroid bed.
! Most solid or mixed thyroid nodules greater than 1 cm should undergo cytologic evalua-
tion before surgery with increasing consideration for universal or selective use of
biomarker assays.
! Biomarkers such as Braf add value to standard cytopathology in identifying suspected
well-differentiated thyroid cancers.
! Biomarkers have prognostic value and with additional confirmatory information may
help decision making regarding extent of surgical treatment and application of adjuvant
treatments.

Disclosures: No disclosures (J. Bumpous); National Institutes of Health grant on PET detector
development, PI ECOG/ACRIN 6685, AO Faculty (B.C. Stack); Consultant, Stryker Corporation
(E. Pribitkin).
a
Division of Otolaryngology-Head and Neck Surgery, University of Louisville, Louisville, KY,
USA; b Department of Otolaryngology-Head and Neck Surgery, Jefferson University College
of Medicine, 925 Chestnut Street, 6th Floor, Philadelphia, PA 19107, USA; c Department of
Otolaryngology-Head and Neck Surgery, University of Arkansas for Medical Sciences, Little
Rock, AR, USA
* Corresponding author.
E-mail address: [email protected]

Otolaryngol Clin N Am 47 (2014) 609–623

http://dx.doi.org/10.1016/j.otc.2014.04.007 oto.theclinics.com
0030-6665/14/$ – see front matter ! 2014 Elsevier Inc. All rights reserved.
610 Bumpous et al

INTRODUCTION

In 2013, in the United States, it was estimated that 60,220 new cases of thyroid cancer
would be diagnosed and 1850 deaths would be caused by thyroid cancer.1 Thyroid
cancer affects women more often than men and usually occurs in people between
the ages of 25 and 65 years.2 The incidence of this malignancy has been increasing
over the last decade.2 Approximately 60,000 thyroid surgeries are performed annually,
of which 33% (20,000) are thyroid lobectomies.3
Thyroid cancer risk factors include a history of radiation, goiter, a family history of
thyroid disease, the female gender, and the Asian race.4 Established clinical prog-
nostic factors in well-differentiated thyroid cancer include age greater than 40 years,
extrathyroidal/extracapsular invasion, vascular invasion, male gender, follicular dis-
ease, and tumors greater than 4 cm. Lymph node status does not seem to affect
disease-free survival.
Risk of a nodule being malignant include size, cold nodule status, ultrasono-
graphic (US) features (microcalcifications and increased nodular vascularity), a
neck radiation exposure history, a family history in 1 or more first-degree relatives,
associated lymphadenopathy on presentation, cytopathology (Bethesda grade)
(Box 1, Table 1), and biomarker results (Afirma [Gene Expression Classifier Vera-
cyte, Inc, San Francisco, CA, USA], MiRInform Thyroid [Asuragen, Inc, Austin, TX,
USA], Thyroseq [University of Pittsburgh, Pittsburgh, PA, USA], microRNA).5,6
Molecular testing is a developing modality to be used judiciously in clinical practice.
Much needs to be studied and reported regarding optimal and cost-effective use of
molecular testing in the context of nodular thyroid disease (Table 2). This article in-
cludes cases that we hope show how molecular biomarker testing of thyroid nodule
fine-needle aspirates (FNA) may be appropriately leveraged in a thyroid surgical prac-
tice (Table 3).

Box 1
The Bethesda system for reporting thyroid cytopathology: recommended diagnostic
categories

I. Nondiagnostic or unsatisfactory
Cyst fluid only
Virtually acellular specimen
Other (eg, obscuring blood, clotting artifact)
II. Benign
Consistent with a benign follicular nodule (includes adenomatoid nodule, colloid
nodule)
Consistent with lymphocytic (Hashimoto) thyroiditis in the proper clinical context
Consistent with granulomatous (subacute) thyroiditis
Other
III. Atypia of undetermined significance or follicular lesion of undetermined significance
IV. Follicular neoplasm or suspicious for a follicular neoplasm
Specify if Hürthle cell (oncocytic) type
V. Suspicious for malignancy
Suspicious for papillary carcinoma
 Algorithms from Experts for Molecular Testing 611

Suspicious for medullary carcinoma

Suspicious for metastatic carcinoma
Suspicious for lymphoma
Other
VI. Malignant
Papillary thyroid carcinoma
Poorly differentiated carcinoma
Medullary thyroid carcinoma
Undifferentiated (anaplastic) carcinoma
Squamous cell carcinoma
Carcinoma with mixed features (specify)
Metastatic carcinoma
Non-Hodgkin lymphoma
Other

From Baloch ZW, Alexander EK, Gharib H, et al. Overview of diagnostic terminology and report-
ing. In: Cibas ES, Ali SZ. The Bethesda system for reporting thyroid cytopathology. New York:
Springer, 2010; with permission.

Table 1
The Bethesda system for reporting thyroid cytopathology: implied risk of malignancy and
recommended clinical management

Risk of
Diagnostic Category Malignancy (%) Usual Managementa
Nondiagnostic or unsatisfactory 1–4 Repeat FNA with US guidance
Benign 0–3 Clinical follow-up
Atypia of undetermined significance or w5–15b Repeat FNA
follicular lesion of undetermined
Significance
Follicular neoplasm or suspicious for a 15–30 Surgical lobectomy
follicular neoplasm
Suspicious for malignancy 60–75 Near-tota1 thyroidectomy or
sugica1 lobectomyc
Malignant 97–99 Near-total thyroidectomyc

Abbreviation: FNA, fine-needle aspiration.

a
Management may depend on other factors (eg, clinical, sonographic) besides the FNA
interpretation.
b
Estimate extrapolated from histopathologic data from patients with repeated atypicals.
c
In the case of suspicious for metastatic tumor or a malignant interpretation indicating metasta-
tic tumor rather than a primary thyroid malignancy, surgery may not be indicated.
From Cibas ES, Ali SZ. The Bethesda system for reporting thyroid cytopathology. Am J Clin Pathol
2009;132:658–65; with permission.
612 Bumpous et al

Table 2
Analysis of thyroid nodule cytopathology process and optimization

Component Description Optimization

Pathology Follicular-derived lesions and Development of tests distinguishing
neoplasms have overlapping the pathobiological nature of
features lesions
Certain neoplasms (eg, classic
papillary carcinoma) have
characteristic features
FNA operator Aspiration technique Cellular specimen reflecting the
Capillary (nonaspiration) technique architectural pattern of cell
proliferation
Specimen Direct smears Slides: monolayer of well-preserved
processing Liquid-based slides and well-stained cells without
Cell block artifacts
Collection in molecular preservative Ancillary studies: preservation of
nucleic acid and protein molecules
Interpretation Bethesda classification (6 tiered) Standardization and application of
and reporting criteria for uniform diagnostic
practice
Ancillary studies Molecular markers: BRAF, RAS, High sensitivity and specificity for
RET/PTC, PAX8/PPAR-g malignant neoplasms
Immunohistochemical markers:
HBME-1, galectin-3, CITED-1
Others under development

From Ohori NP, Schoedel KE. Thyroid cytology: challenges in the pursuit of low-grade malig-
nancies. Radiol Clin North Am 2011;49:435–51; with permission.

Table 3
Bethesda classification of FNA compared with the usefulness of currently available molecular
marker testing

Thyroid Cytology
Classes of FNA Results Usefulness of Molecular Testing
Negative (II) Not useful
Rule out follicular neoplasm (III) Useful
Follicular neoplasm (IV) Most useful
Suspicious for malignancy (V) Probably useful (30%–40% still benign at pathology)
Positive (VI) Not useful

CASES FOR DISCUSSION

Case 1: Incidentally Noted Thyroid Nodule on Carotid Ultrasound
! A 54-year-old man presented with an incidental 1.5-cm mixed cystic solid nodule
noted on carotid vascular US. Diagnostic US-guided FNA biopsy (FNAB) was
performed. The interpretation was benign colloid goiter, but the case was re-
viewed by a second pathologist, who made the following comment: “If malig-
nancy is suspected in this solitary thyroid follicular lesion, malignancy cannot
be definitively excluded until the questionable mass is completely excised and
examined histologically. Clinical correlation is required to determine the next
course of action.”
 Algorithms from Experts for Molecular Testing 613

Bumpous
The context of this case is a very familiar scenario: a nodule incidentally noted dur-
ing a diagnostic carotid vascular study. On the US (Fig. 1), you can see some of the
features that are classic: this nodule meets size criteria (>1 cm), is heterogeneous
with cystic and solid features, and merits an FNAB. As you look at the Doppler
flow, it is mostly a peripheral pattern, which points toward a benign process; how-
ever, the patient underwent an US-guided aspiration biopsy, which showed a follic-
ular aspirate with some blood and colloid. We now get into a situation that I think
this case frames nicely. Pathology’s statement here is that a malignancy is sus-
pected and it cannot be definitively excluded. By a disclaimer at the bottom of
the report, you get the impression that it is benign, but there is no certainty as to

Fig. 1. (A) 1.2-cm by 1.0-cm left thyroid lobe lesion, complex cyst, axial view. (B) Same lesion
as (A) in axial view with the Doppler setting on showing peripheral vascularity.
614 Bumpous et al

what it is. So, do you proceed with lobectomy, which is the most aggressive
approach, or do you repeat the FNA? If you could increase certainty, do you include
molecular panels for evaluation of the nodule, or is this a patient who you have a
conversation with and simply feel comfortable observing? I think all of these options
may have a role depending on the clinical context. If you were doing molecular
testing, you would have an informed conversation with the patient and involve the
patient in the decision making. What could be the results of the test? How would
that influence us? Who pays for the test?

Pribitkin
This is a very common scenario, in which the pathologist does not translate the
descriptive diagnosis into the corresponding Bethesda classification category. I would
say that this is a follicular lesion of undetermined significance (FLUS). The key here is
determining the prevalence of malignancy in your patient population with FLUS on
FNAB. At Jefferson, we can tell the patient with this FLUS cytology that about 15%
of patients with a similar cytology will have a final diagnosis of cancer. I also tell the
patient that almost all these cancers are low grade, slowly increasing in size, not likely
to cause death or significant morbidity. In other series of patients featuring a 25% to
40% prevalence of malignancy in FLUS cytologies, it is more reasonable to recom-
mend a thyroid lobectomy. At Jefferson, we would recommend that this patient un-
dergo a repeat US-guided FNAB with molecular panel testing.

Stack
Speaking to the point that Dr Pribitkin made regarding the prevalence of malignancy in
your institution, when these diagnoses are made by your pathologist, one might ask,
“Well how do you do that?” Random conversations between the surgeon and the
pathologist may never arrive at any conclusive data. Our approach at the University
of Arkansas for Medical Sciences is to hold a monthly meeting among the surgeons,
pathologists, and the radiologists. It is our equivalent of a thyroid cancer tumor board,
and we review the pathology of all the cases. Now, some of the pathology will be
cytology and will be presented before the surgery. Often, these cases are evaluated
retrospectively, so we have the benefit of having cytology and then having operative
findings, which include the final pathology. We are able to present these together
and engage in conversations that refine our approach to cancer diagnosis. I believe
this an important way that you can reach this understanding of the prevalence of
malignancy within the various Bethesda classifications of cytology, given the various
different interpretations that may come from your pathologists from within your
institution.

Bumpous
The tumor board is an outstanding idea. We do the same thing. When you have a high
volume, you are able to make those kinds of conclusions. If you do not have a high
volume, you have to look at the data throughout your area and look at what leading
practices are doing in your area at the university centers. Let us say it is a 15% inci-
dence of a malignancy in FLUS in your center, then I think you can either use the
Afirma MiRInform Thyroid, or Thyroseq tests. All have a relatively high negative predic-
tive value at that level of prevalence. When you go up a little higher to a 25% preva-
lence of malignancy in your FLUS patients, then MiRInform Thyroid demonstrates a
diminishing negative predictive value, but both Afirma and Thyroseq hold up in the
terms of the negative predictive value. So, in that situation, you would favor doing
the Afirma or Thyroseq test as a rule-out negative predictive value test.
 Algorithms from Experts for Molecular Testing 615

Case 2: Asymptomatic Thyroid Mass

! A 50-year-old woman presented with a several-year history of a right thyroid mass.
She complained of no symptoms such as dysphagia or airway compromise. The
lesion had recently been noticed by a new primary care provider, and referral was
made for evaluation and management. On evaluation, the patient had a 2-cm
palpable right thyroid mass. The remaining examination was normal. The vocal
folds were completely functional on examination. US was performed and showed
a 2.6-cm right thyroid lesion, which was heterogeneous without microcalcifica-
tions. There were flow patterns around the lesion suggestive of thyroiditis.

Stack
Our concern here is not necessarily of a change in the clinical situation triggering con-
cerns about thyroid lymphoma or anaplastic carcinoma, but this is a long-standing
problem, which perhaps has been neglected by the patient or by the primary care doc-
tor. When the patient is evaluated more thoroughly, it reveals a 2-cm lesion in greatest
diameter by US, which shows some concern for malignancy. There is also another
very common finding of altered vascular flow. Certainly, in middle-aged and older fe-
males, vascular flow suggestive of thyroiditis is endemic in North America. We see the
cytology both by low power and high power (Fig. 2). We have a follicular lesion, and so
cytology on its own is not able to distinguish between a follicular carcinoma and a
follicular adenoma; perhaps we are just looking at an abundance of follicular cells,
and it is not a follicular lesion but just follicular cells. So, there are limits to what
cytology can give to us in our operative decision making. This would be another
case in which molecular testing could be deployed. In our institution, we have 3 excel-
lent thyroid cytopathologists, and we have not been offering molecular testing to these
patients, although it would be appropriate to do so here. Likely, we would repeat the
needle biopsy, and if it again showed a predominance of follicular cells and was sug-
gestive of follicular lesion, we would likely proceed to an outpatient diagnostic hemi-
thyroidectomy. Now, if we are going to use surgery as part of making a diagnosis, then
the next logical question is when do you use frozen sections? Largely, we do not use
frozen sections to make decisions regarding proceeding to total thyroidectomy. Occa-
sionally, if we are concerned with papillary cancer, we do a frozen section on a lymph
node, but rarely on the thyroid specimen, unless there are clinical indications of
aggressive behavior, such as extrathyroidal extension. Frozen sections do not allow
for the comprehensive evaluation of a follicular lesion to determine if there are areas
where there is vascular or capsular invasion.

Fig. 2. (A) FNA of a follicular at low power. (B) FNA of a follicular at high power. (Papani-
coloau stain, A&B).
616 Bumpous et al

Bumpous
This is another common scenario, in which you have to exercise some cautions; in
particular, I think that the image guidance of the biopsy by an experienced ultraso-
nographer is critical. If you have a situation where you are suspicious of thyroiditis
and a follicular process, you are much more likely to obtain follicular features, and
those might be coexistent with the other disease. In frozen section, standardizations
of processes are very important; in particular, the number of sections taken through
a lesion. Simply bivalving the nodule in hopes of determining that this is a follicular
carcinoma versus follicular adenoma is unlikely to be a sufficient examination to
make that determination. Studies have been quite good at looking at frozen sec-
tions, and if you increase the number of sections that are actually done with the
nodule, the security of developing that diagnosis is much better. So, there are com-
petencies and processes within pathology which are critical to make frozen section
valuable. In my institution, I use frozen sections on a selective basis usually to
confirm a clinical suspicion. If I see, for example, fibrous changes or if I am suspi-
cious of muscle invasion, or if I see a lymph node that appears suspicious, that is
the context that I clinically consider frozen sections as a support for increasing
the extent of surgery.

Pribitkin
I would agree that you need to have the processes in place. I think it is an interesting
question in terms of your frozen section. We looked at this many years ago before the
era of molecular testing, and what we found was that intraoperative cytology, a touch
prep, of the lesion during the procedure was most effective in terms of making a
diagnosis. I would urge you to speak with your pathologist to make sure that cytology
is done. When we looked at the cost-effectiveness of intraoperative frozen section,
we also discovered the reduction in costs arose not from the identification of follic-
ular carcinomas, but rather from the identification of follicular variants of papillary
thyroid cancer (FVPTC). At that time, these intraoperative FVPTC diagnoses promp-
ted us to do a total thyroidectomy and saved the patient another operation, therefore
making intraoperative cytology cost effective. Interestingly, the FNAB data including
preoperative testing with molecular diagnosis show that BRAF and NRAS testing
limits the cost-effectiveness of intraoperative cytology. Essentially, you obviate
frozen sections by incorporating molecular testing results into your preoperative de-
cision making.

Case 3: Rapidly Enlarging Thyroid Mass

! A 30-year-old man presented with left thyroid mass that had been rapidly
growing over 3 months. He self-referred because he was distressed about
the possibility of cancer. Physical examination was a healthy young man with a
3-cm left thyroid mass, which was tender to palpation. US and FNA were per-
formed, and 2 lesions were noted. The 3-cm lesion was cystic; its cytology is
shown in Fig. 3A. There was an adjacent 7-mm left lobe lesion, which was solid
and hypervascular; its cytology is shown in Fig. 3B.

Stack
Unlike our previous case, this is a new finding clinically and, therefore, raises the
concern on the part of the patient and the providers as to whether there is cancer.
A patient may come in with 1 clinical finding, and others are found on US. The actual
lesion that prompted presentation had benign cytology with a very classic appearance
where the colloid has this bubble gum pink/purple appearance spread across the slide
 Algorithms from Experts for Molecular Testing 617

Fig. 3. (A) Bubble gum colloid appearance of a benign colloid nodule aspirate. (B) FNA of a
Hürthle cell lesion. (Papanicoloau stain, A&B).

(see Fig. 3A). The incidentally found lesion adjacent to the dominant lesion is concern-
ing for a Hürthle cell neoplasm. Again, we have a similar quandary: what options would
we offer for the patient? Is this a patient for whom molecular testing would be appro-
priate? And would you base your surgical decisions on intraoperative frozen sections?
So I am very interested to hear what my colleagues are going to say about this,
because this is very similar to our previous case with regards to whether we should
offer observation or lobectomy with or without frozen section.

Pribitkin
Seven-millimeter lesions are relatively small, and this is an incidental finding, because,
if you were not biopsying the big lesion, you would not have biopsied the 7-mm lesion,
unless it had some suspicious characteristics on US. I think you can reassure the pa-
tient to some extent by offering molecular testing. The molecular testing would help us
to decide whether or not a lobectomy should be done. Given the small size of the
lesion, a lobectomy should be sufficient, unless the final pathology revealed an un-
usual malignancy.

Stack
I appreciate the remarks of Dr Pribitkin, and let me just plays devil’s advocate here.
You have a lesion that by definition could be a carcinoma. If it was a carcinoma,
and you could offer him a thyroidectomy, which would be both a diagnostic and ther-
apeutic given a totally clean contralateral lobe, Dr Bumpous, would this change your
approach? Would you forgo perhaps molecular analysis since the final common
pathway seems to be coming together? Would you not bother with frozen sections
in this particular case but just defer to permanent pathology?

Bumpous
This is a scenario where you have an intersection of science and medicine, and it un-
derscores the importance of informed conversations with your patients. The likelihood
of this being malignant is fairly low and, even if it were malignancy, the implications of it
are relatively small. On the other hand, if it were an 80-year-old patient, you would feel
much more comfortable proceeding perhaps with a more conservative approach. I do
not think it is unreasonable to perform a lobectomy on this patient. What you have to
inform the patient of in this situation is if there could be findings that might indicate that
this was more aggressive. If you have multifocal carcinoma that is found at the time of
permanent pathology, maybe this is someone who may have to consider a completion
of thyroidectomy. I think that is unlikely given the fact that you have pretty good US
618 Bumpous et al

examination. You have to provide the patient with the options of repeat FNA or initial
FNA with molecular testing. I still think it is reasonable to perform the lobectomy.
I agree with Dr Pribitkin’s comments earlier, and I do not think proceeding with total
thyroidectomy is the first step.

Pribitkin
One of the things to look for on final pathology, given that a lobectomy is done, is a
Hürthle cell or follicular malignancy. We have to distinguish between a minimally inva-
sive cancer, where there is just invasion of the capsule, and a cancer with vascular or
lymphatic invasion. In the former case, a lobectomy is sufficient. But a Hürthle cell can-
cer that has vascular invasion, even if it is a small lesion, is at risk for more distant
spread. In those situations, we would recommend a total thyroidectomy, so that we
can more easily screen for distant metastases with serum thyroglobulin and better
follow the patient long-term.

Stack
I appreciate Dr Pribitkin’s remarks as far as the surgery is concerned, and in the event
that this came back Hürthle cell carcinoma with signs of vascular invasion, I totally
concur with completion of total thyroidectomy. My question for Dr Pribitkin is that,
given that we know the size of this lesion, and that we have no other surprise findings
on final pathology on the left lobe, would he consider doing a left central neck dissec-
tion (CND), because of the findings of vascular invasion.

Pribitkin
No, we would not. There are no signs of lymph node enlargement on US or on other
types of imaging, and Hürthle cell carcinomas are more likely to spread hematoge-
nously. We would not, therefore, do a CND, especially, given the increased morbidity
of a CND performed in conjunction with a completion thyroidectomy 1 to 2 weeks
following the initial surgery. If the thyroglobulin remained elevated postoperatively,
we would obtain both an iodine radionuclide and a positron emission tomography
computed tomography scan to evaluate the patient for distant metastases.

Case 4: Incidentally Discovered Thyroid Mass During Carotid Artery Evaluation

! A 60-year-old man presented with a left thyroid mass incidentally discovered on
a vascular duplex study for carotid artery evaluation. His physical examination
was normal, and the lesion was not palpable to your examination. A US exami-
nation showed a 1.4-cm left thyroid lesion in a background of thyroiditis. There
was some increased perilesional vascular flow, and a few microcalcifications
were appreciated. An FNA was performed and returned as in Fig. 4. In counseling
the patient, what advice do you give? He asked if there was a role for molecular
testing and you said no. Why?

Stack
Another challenge is when older patients come in with incidental findings, whether it is
through PET scanning for surveillance of another cancer or, in this case, a carotid
scan. There is some concern about carotid artery disease, and when performing
this examination, US reveals a 1.4-cm left thyroid lesion with some concerning find-
ings: there appears to be some increase of vascular flow around the lesion, which is
a soft concerning sign that the lesion is malignant. The FNA is performed appropri-
ately, and it returns findings which are classic for papillary thyroid cancer (PTC).
What advice would you give to the patient? And given that there appears to be
some certainty in the diagnosis, do we need to do molecular testing in this case?
 Algorithms from Experts for Molecular Testing 619

Fig. 4. FNA positive for papillary thyroid cancer. (Papanicoloau stain).

Bumpous
When you have a fairly clear situation, molecular testing does not offer a considerable
amount of added value, and it does add cost. There is also some trepidation that I
have, as we have a case here that looks very classic: nuclear grooves, papillary
type of pathology that is diagnostic for carcinoma. If you had testing that was counter
to the diagnosis of papillary carcinoma, would that persuade you from doing at least a
lobectomy on this patient, given these cytologic findings? I think in this situation,
molecular testing is redundant and could confuse the clinical decision making.

Pribitkin
I agree that this patient requires a total thyroidectomy. If you have confidence in your
cytopathology, then there really is no need for intraoperative frozen section. I do not
think that you can be faulted, if you do have some concerns about the cytopathology,
to get an intraoperative frozen section to increase your level of confidence. Occasion-
ally, there are scenarios where you operate on 1 side of the thyroid and you run into
difficulties. If you run into difficulties on 1 side, then, before you go to the other
side, get frozen section pathology to confirm the diagnosis, and you may avoid doing
a total thyroidectomy in an otherwise very difficult situation. Given that this is a small
cancer in an elderly gentleman and the question arises, in what situations would you
perform a CND?

Stack
This patient is a clinically N0 neck based on the information that we are given here.
With a thyroid gland in situ in the neck, it is difficult to do a comprehensive US evalu-
ation of level 6 to know that there is adenopathy. You have 1 further chance, and that is
during the removal of the thyroid, you may find some clinical adenopathy, and again,
we are not given that information on this particular case. But finally, we have the
American Thyroid Association (ATA) guidelines that we can look to for advice. The
ATA guidelines do not recommend routine (prophylactic) CND for T1 and T2 lesions,
which would be the category that this lesion falls into. So, absent finding nodes while
doing the thyroidectomy, I would not go ahead and do a CND.

Bumpous
I agree wholeheartedly with Dr Stack. I would not routinely perform a CND in this sit-
uation. There are some findings that would upstage the patient, such as intraoperative
findings of capsular invasion or clinically positive lymphadenopathy. When you have
620 Bumpous et al

an intermediate-size lymph node, it is again an area where frozen section can assist in
your decision, because identifying a papillary carcinoma within the lymph node is a
relatively straightforward task for most pathologists. Frozen section analysis of the
lymph node can help with the decision about expanding the operation. I always inform
the patient preoperatively that there is the possibility that a limited lymph node dissec-
tion may be required. I think another issue that is also important is the extent of the
central compartment neck dissection. Do you remove all of the nodes that are lateral
to the recurrent nerve or can you restrict your dissection to those nodes that are medial
to the recurrent nerve? What are the node groupings within the central compartment
that need to be assessed, or do all of them routinely need to be assessed? We would
be remiss not to acknowledge that the incidence of temporary and permanent hypo-
parathyroidism is increased in the face of central compartment nodal dissection.

Case 5: Thyroid Mass Post Removal of Parathyroid Adenoma

! A 50-year-old female patient who had had removal of left inferior parathyroid ad-
enoma 1 year previously presented with US showing a right 1-cm thyroid mass
with microcalcifications. There was no lymphadenopathy or contralateral thyroid
nodules. There was no history of neck irradiation or family history of thyroid
cancer. The patient was euthyroid, and the hyperparathyroidism was cured
with unilateral exploration. The FNA was suspicious for PTC; BRAF and NRAS
were negative (Fig. 5A, B).
! The patient underwent right thyroid lobectomy. No lymph nodes were encoun-
tered at surgery. Both parathyroids were preserved. Final pathology showed

Fig. 5. (A) Right thyroid lesion in transverse (axial) view. (B) Same lesion as (A) in a sagittal
view. (C) Low-power hematoxylin-eosin view of same lesion as (A), Arrow area is magnified
in D. (D) High-power hematoxylin-eosin view of same lesion as (A), PTC.
 Algorithms from Experts for Molecular Testing 621

8-mm PTC, with 1 focus of extrathyroidal extension of PTC. There were no para-
thyroids in the specimen (see Fig. 5C, D).

Stack
It is important that when you see characteristics that are suspicious on US, it raises
your suspicion for doing FNA and ultimately recommending surgery, even if the aspi-
ration is equivocal. Microcalcifications as well as irregular borders on US are worri-
some findings. These findings move you up that level to be concerned for a
malignancy. Again, no lymphadenopathy was seen on US. When we look at both
the longitudinal and the transverse view, we can also get an appreciation of whether
the lesion is taller than it is wide, which is also associated with an increasing likelihood
of malignancy. The FNA was suspicious for papillary thyroid carcinoma and there was
molecular testing done. The molecular testing showed both BRAF and NRAS negative.
So the more common molecular tests for malignancy were negative but the cytopa-
thology was suspicious for PTC. Here, we have a lesion that is suspicious but not
conclusive papillary thyroid carcinoma. So, what would the recommendation be in
terms of initial surgery?

Bumpous
This situation is one in which you have disparate data and you have US findings and
cytopathologic findings that are highly suspicious for malignancy, yet your molecular
testing is not pointing in that direction. There are a variety of reasons that this can
occur. I think that this is a situation in which you exercise caution and have a conver-
sation with the patient: we have features that are pointing to malignancy but not a
definitive diagnosis of malignancy, so in my estimation, this is a perfect case for a thy-
roid lobectomy. With an assessment of either frozen section or permanent section,
you have high rates of confidence and good quality control. With frozen section,
you may or may not be able to make intraoperative decision making, based on previ-
ously discussed parameters.

Stack
In my particular setting, given this lesion has about 60% chance of being PTC. I would
start with a thyroid lobectomy. I may or may not, based on the clinical situation, do a
frozen, but I would in all likelihood proceed to doing total thyroidectomy at this oper-
ative setting. Given the aggressive final pathology, if only a thyroid lobectomy were
performed, I would proceed with a completion thyroidectomy. When I do completion
thyroid surgery, I assume that there are no functioning parathyroids on the previously
operative side, whether it is myself or another surgeon who referred the case. And the
fact that the patient had at least 1 parathyroid removed previously does not dissuade
me from this more aggressive thyroidectomy posture.

Pribitkin
The 1 bit of data that exists in some of the larger studies, but has not been conclusively
proven, is that BRAF-negative low-risk microcarcinomas cases exhibit a far lower
mortality and disease recurrence. There is a suggestion that a lobectomy could poten-
tially suffice in these patients, especially if there is concern about a higher risk of
having a problem on the other side during the surgery. This patient did undergo just
the right thyroid lobectomy, there were no lymph nodes encountered in surgery, every-
thing went very smoothly, and the final pathology showed a microcarcinoma (8 mm),
but it did show extrathyroidal extension. The question now is whether one proceeds
with a completion thyroidectomy for this microcarcinoma with a little bit of extrathyr-
oidal extension? Is a central neck indicated as well?
622 Bumpous et al

Bumpous
The extrathyroidal extension is a finding you should not ignore. I would proceed with a
completion thyroidectomy and at least an ipsilateral CND on the side of the lesion. The
rationale for doing this is that there is a high rate of a lymph node metastasis, so a CND
provides further pathologic staging that would inform decisions regarding adjuvant
therapy.
Stack
There are 3 reasons why we want to do a total thyroidectomy: number 1 is that these
lesions can be multifocal, and multifocality is both a prognostic and a therapeutic issue
in treating thyroid cancer. There is also the issue that remaining thyroid tissue will be
an impediment to radioactive iodine, and finally, the use of serum thyroglobulin is not
effective, unless a total thyroidectomy has been performed. These are 3 reasons in my
mind, in addition to guideline reasons, in addition to the diagnosis of cancer, that we
would do completion thyroidectomy. I would concur with the need to do unilateral
CND, because now we know that with final pathology, this is at least a T3 lesion
with minimal extrathyroidal spread, and it would be a T4 with more extensive extra-
thyroidal spread.
Case 6: Incidentally Discovered 2.5 cm Nodule
! A 33-year-old patient presents with an incidentally found right 2.5-cm nodule
(Fig. 6A). There was no lymphadenopathy or contralateral thyroid nodule. There
was no history of neck irradiation or a family history thyroid cancer. The patient
was euthyroid. FNA showed a follicular neoplasm; BRAF was negative but
NRAS positive.

Pribitkin
This is a scenario in which the suspicion for a malignant follicular neoplasm is much
higher, and I would offer the patient lobectomy, with possibly a confirmatory frozen
section (see Fig. 6B). I think one of the issues here is the value of the upfront molecular
testing. When we get an NRAS-positive result, there have been some centers that
have suggested that this carries an 87% incidence of a malignancy of some type.
The most common malignancy is a follicular variant of PTC. Some centers would pro-
ceed with a total thyroidectomy, because at the University of Pittsburgh, those individ-
uals who had a follicular variant of papillary cancer also had an additional 50% risk of
having a contralateral follicular variant of papillary cancer, even in the presence of a
normal lobe on US.

Fig. 6. (A) A 2.5-cm right thyroid nodule. (B) High-power view of same lesion as (A), follic-
ular neoplasm. (hematoxylin-eosin).
 Algorithms from Experts for Molecular Testing 623

Stack
In my practice, we would not be using the molecular testing with any regularity. We
would not have that information. We already have the cytology, which raises our index
of suspicion that we may be dealing with a malignancy. In this particular situation, I
would do a frozen section. Intraoperative frozen sections can help to identify papillary
cancer. And based on that information, I would proceed to do a total thyroidectomy, I
would do an intraoperative examination of level 6 and do any lymph node dissection
that would be indicated based on my intraoperative findings.

REFERENCES

1. American Cancer Society. Cancer facts and figures 2013. Atlanta (GA): American
Cancer Society; 2013.
2. Tennvall J, Biörklund A, Möller T, et al. Is the EORTC prognostic index of thyroid
cancer valid in differentiated thyroid carcinoma? Retrospective multivariate anal-
ysis of differentiated thyroid carcinoma with long follow-up. Cancer 1986;57(7):
1405–14.
3. Vashishta R, Mahalingam-Dhingra A, Lander L, et al. Thyroidectomy outcomes: a
national perspective. Otolaryngol Head Neck Surg 2012;147(6):1027–34.
4. Iribarren C, Haselkorn T, Tekawa IS, et al. Cohort study of thyroid cancer in a San
Francisco Bay area population. Int J Cancer 2001;93(5):745–50.
5. Nikiforova MN, Wald AI, Roy S, et al. Targeted next-generation sequencing panel
(ThyroSeq) for detection of mutations in thyroid cancer. J Clin Endocrinol Metab
2013;98(11):E1852–60. http://dx.doi.org/10.1210/jc.2013-2292.
6. Gupta N, Dasyam AK, Carty SE, et al. RAS mutations in thyroid FNA specimens
are highly predictive of predominantly low-risk follicular-pattern cancers. J Clin
Endocrinol Metab 2013;98(5):E914–22. http://dx.doi.org/10.1210/jc.2012-3396.