Anaplastic thyroid carcinoma: A 50-year

experience at a single institution

Bryan McIver, MB, PhD, Ian D. Hay, MB, PhD, Dario F. Giuffrida, MD, Catherine E. Dvorak, RN,
Clive S. Grant, MD, Geoffrey B. Thompson, MD, Jon A. van Heerden, MD, and John R. Goellner, MD,
Rochester, Minn

Background. Anaplastic thyroid carcinoma (ATC) is among the most aggressive of human malignan-
cies. However, there have been few large studies of histologically well-defined ATC. We report the results
of a 50-year experience of this lethal malignancy.
Methods. We reviewed all cases of ATC managed in this institution between 1949 and 1999.
One pathologist (J.R.G.) reviewed all pathologic material. Clinical details were obtained from medical
records, and current status of all patients was determined.
Results. There were 134 cases, with a female-to-male ratio of 1.5:1 and a mean age of 67 years. Benign
thyroid disease was present in 27 cases (20%) and well-differentiated thyroid carcinoma in 31 (23%).
Sixty-two patients (46%) had distant metastases at diagnosis, and 98% of the tumors were locally
invasive. Primary treatment was surgical for 96 patients (72%). Complete resection was achieved in
29 cases (30%), with “minimal residual disease” in 25. Neither extent of operation nor completeness of
resection affected survival ( P > .4). Postoperative radiotherapy gave slightly longer median survival
(5 vs 3 months), which was not significant ( P < .08). Multimodal therapy, including operation,
chemotherapy, and radiotherapy, did not improve survival.
Conclusions. The outlook for patients with ATC remains grim. Novel treatments for ATC are desperately
needed. (Surgery 2001;130:1028-34.)

From the Departments of Medicine, Surgery and Surgical Pathology, Mayo Clinic & Foundation, Rochester,
Minn

ANAPLASTIC THYROID CARCINOMA (ATC) is among
the most aggressive malignancies and stands in
stark contrast to the usually indolent behavior of
differentiated thyroid carcinoma. Although ATC
represents only 2% to 5% of clinically recognized
cancers of the thyroid gland,1 more than half of
the approximately 1200 deaths attributed to thy-
roid cancer annually in the United States result
from ATC. Mortality rates of 70% to 95% have
been reported, but a small proportion of patients
experience long-term survival.2,3
ATC includes large-cell, spindle-cell, and
small-cell variants.4 However, recent advances in
immunohistology have shown that most of the
small-cell tumors are actually non-Hodgkin’s lym-
phoma of the thyroid, medullary carcinoma, or
Supported in part by the Mayo Foundation Endocrine
Neoplasia Program.
Presented at the 22nd Annual Meeting of the American
Association of Endocrine Surgeons, Atlanta, Ga, April 28-May 1,
2001.
Reprint requests: Bryan McIver, MB, PhD, Division of
Endocrinology, Mayo Clinic, 200 First St SW, Rochester, MN
55905.
Copyright © 2001 by Mosby, Inc.
0039-6060/2001/$35.00 + 0 11/6/118266
doi:10.1067/msy.2001.118266
“insular” variants of follicular carcinoma, all of
which may carry a better prognosis than ATC.5 Few
large studies have been reported of histologically
well-defined ATC, based on current histologic cri-
teria. Our understanding of prognostic variables
remains poor, and the roles of operation, radiation
therapy, and chemotherapy are uncertain, al-
though a number of small studies have reported
improved outcomes after multimodal therapy.6-8
We report the results of a large retrospective
review of patients managed at a single institution
for a diagnosis of ATC during a 50-year period. In
all cases, the histology was reviewed and the diag-
nosis was confirmed by current criteria. The clini-
cal presentation, association with pre-existing
benign and malignant thyroid disease, outcomes,
and impact of treatment are presented for this
almost universally lethal malignancy.
METHODS
These studies were approved by the In-
stitutional Review Board. We identified all cases of
anaplastic, undifferentiated, or poorly differentiat-
ed thyroid carcinoma (DTC) presenting to this
institution between 1949 and 1999. Patients were
excluded if they had received prior treatment for
this diagnosis or if more than 30 days had passed

1028 SURGERY
Surgery
Volume 130, Number 6
Fig 1. Cases of ATC presenting to the Institution in each
5-year interval between 1949 and 1999. An average of 2.7
cases per year were seen during this period, with no sig-
nificant change during 50 years.
since the diagnosis was made. One pathologist
(J.R.G.) reviewed the original pathology material,
and ATC was diagnosed by current histologic crite-
ria,9 including large or spindle-shaped cells, cellu-
lar and nuclear atypia, high mitotic rate, and
necrosis. Immunohistochemistry for thyroglobulin
and cytokeratin was performed, particularly in
small-cell lesions, to exclude lymphoma, whereas
staining for calcitonin and amyloid excluded
medullary carcinoma. Cases with clear squamous
differentiation were also excluded.
Clinical, surgical, pathologic, and follow-up
details were obtained from the medical record. We
reviewed death certificates and autopsy reports
when available. A postal survey determined the sta-
tus of patients who remained alive at the time of
last contact. Local recurrence was identified clini-
cally, and distant metastases were diagnosed
radiologically, without requiring histologic con-
firmation, which rarely was obtained. We attributed
death to ATC if there was evidence of progressive
disease in the patient’s terminal illness and no
other proximate cause of death was identified.
We used a software package (SAS Institute Inc,
Cary, NC) for statistical analysis. Differences
between groups were assessed with a chi-square test
and survival with life-table analysis. A P value of .05
or less was considered statistically significant.
RESULTS
ATC was confirmed in 134 cases, and all patients
were followed up to either their death or to
December 1999, a median of 3 months. The mean
of 2.7 cases per year showed no change during 5
decades (Fig 1). Of the 134 patients, 80 (60%)
were women and 54 (40%) were men, a female to
male ratio of 1.5:1. The mean age was 67 years, and
McIver et al 1029
Fig 2. Age distribution of ATC cases during the 50-year
interval. The mean age was 67 years, but 25% of the
patients were younger than 60 at the time of diagnosis. No
change in either age or sex distribution was seen during
the study period.
the age distribution is shown in Fig 2. No change in
the age or sex distribution was seen during 50 years
(data not shown).
Pre-existing thyroid disease. Six patients (4.5%)
had a prior history of DTC. All had been treated
with operation and postoperative radioactive
iodine. In 5 of the 6 cases, the original DTC histol-
ogy was available and showed papillary carcinoma
(PTC) in all 5. The remaining case had been
reported as follicular carcinoma (FTC). ATC was
diagnosed in these patients between 2 and 32 years
after the diagnosis of DTC (mean, 9.5 years). An
additional 25 patients had coincidentally noted
DTC on examination of surgically resected ATC; 17
of these (68%) had PTC, and the remainder had
FTC. Overall, 31 (23%) patients showed evidence
of pre-existing or co-existing DTC, and 22 (71%) of
them had PTC. An additional 27 (20%) patients
had recognized (n = 8) and previously unrecog-
nized (n = 19) thyroid nodules or goiter. Overall,
pre-existent or co-existent thyroid disease was seen
in 58 patients (43%).
Clinical presentation and diagnosis. Almost all
patients (n = 130; 97%) presented with a rapidly
enlarging neck mass, and the remainder were diag-
nosed on the basis of either pulmonary (n = 3) or
skeletal (n = 1) metastases. Sixty-two patients
(46%) had evidence of metastases at diagnosis.
These included pulmonary or mediastinal metas-
tases (56 of the 133 in whom chest imaging was per-
formed), skeletal metastases (9 of 28 in whom bone
scanning was performed), or brain metastases (1 of
11 in whom head computed tomography scanning
was performed). Distant metastases were detected
at some stage of the illness in 91 patients (68%).
Pathology. Large-cell ATC was seen in 38%, spin-
dle cell in 40%, and a mixed picture was seen in 22%.
1030 McIver et al
Fig 3. Surgical procedures undertaken in 83 patients,
whose initial primary therapy was surgical and whose dis-
ease was considered “operable” at the time of operation.
Procedures ranged from lobectomy with tumor debulking
to extensive complex operation with curative intent by
radical or modified-radical neck dissection.
The primary tumor averaged 6.6 cm (range, 1.5-16
cm), and local invasion was evident in 131 (98%)
patients at diagnosis. In all cases for which docu-
mentation was adequate, regional lymph nodes with-
in the neck were involved, often massively so.
Initial therapy. Five patients (4%) were treated
palliatively. External beam irradiation (XRT) was
administered after open biopsy in 29 patients
(22%), and chemotherapy was administered in 4
(3%) patients. Operation was the primary manage-
ment in 96 cases (72%), with 48 (50%) patients
undergoing “debulking” procedures and 35 (36%)
an intended “curative” procedure. Operation was
terminated after biopsy alone in the remaining 13
cases (14%) because of extensive local infiltration.
In patients who had a surgical procedure that was
more extensive than biopsy alone (n = 83), the
procedures ranged from lobectomy to total
thyroidectomy with complex neck dissection. The
distribution of these procedures is shown in Fig 3.
Lymph-node dissection was variable, but most
patients had extensive cervical lymphadenopathy.
The extent of this disease and massive soft-tissue
invasion were the most commonly documented fac-
tors preventing complete surgical resection. The
surgical approach varied during 5 decades, as
shown in Fig 4. During the 1960s, most patients
underwent biopsy alone, followed by XRT as pri-
mary therapy. In contrast, during the 1970s, all
patients underwent operation as primary treat-
ment. In recent decades, operation was used for
most patients, whereas primary XRT was reserved
for massive disease, for which an operation was
believed to carry excessive risk.
Complete resection of the tumor was achieved
in 29 of the surgically treated patients (30%),
Surgery
December 2001
Fig 4. Trends in the surgical approach to ATC. Radical
operation with curative intent and extensive debulking
procedures were used in all patients in the 1970s, whereas
in the 1960s the most commonly used approach was biop-
sy only, in many cases with tracheostomy placement, fol-
lowed by external beam irradiation.
whereas a further 25 (26%) were documented to
have only “minimal residual disease” after the pri-
mary surgical procedure. Gross residual disease was
present after operation in 42 patients (44%).
Subsequent therapy. Postoperatively, 41 of the 42
patients with gross residual disease (98%) and 38 of
the 54 patients with total or near-total resection of
the tumor (70%) received XRT for the primary
tumor. XRT was also used to treat local recurrences
(n = 5) and metastatic disease (n = 4). However, a
local recurrence or metastatic deposit was demon-
strated to decrease in size in only a single patient,
and XRT was administered largely as end-of-life pal-
liation, with no further assessment undertaken in
most patients.
Chemotherapy alone was administered to 4
patients (3%) and as postoperative adjunctive ther-
apy for the treatment of distant metastases in 12
(9%). In the last 20 years, 13 patients have received
“combination therapy” with a debulking (some-
times extensive) procedure, XRT, and adjunctive
chemotherapy (doxorubicin [Adriamycin]) for
“radiosensitization.”
Outcomes and follow-up. The median survival
was 3 months, with 32 patients (23%) surviving less
than 1 month from the time of diagnosis (Fig 5).
Death was attributed to ATC in 127 of the 131
patients who have died (97%), with myocardial
infarction and stroke causing the other 4 deaths.
Thirteen patients (9.7%) have survived more than 1
year from diagnosis. The longest-surviving patient
remains alive with no evidence of disease 23 years
after the diagnosis and treatment of ATC. This
patient, aged 50, had a 6.1-cm intrathyroidal prima-
ry, with 3 small lymph nodes involved, and no evi-
dence of distant metastases (pT3 N1 M0) and was
Surgery
Volume 130, Number 6
Fig 5. Overall survival after a diagnosis of ATC. The medi-
an survival was 3 months from diagnosis, with only 13
patients (9.7%) surviving longer than 1 year. Cause-specif-
ic figures are indistinguishable from this data.
treated with operation alone (lobectomy) with com-
plete tumor resection. No postoperative radiation or
chemotherapy was administered. Of the other 12
“long-term” survivors (> 1 year), all were treated sur-
gically; complete tumor resection was achieved in 7,
and only minimal residual disease was reported in 3.
Both operation and radiotherapy improved sur-
vival over palliative treatment alone, with patients
who received only palliative care having a median
survival of 3 weeks compared with 2.3 months for
patients treated with radiotherapy (P < .005) and
3.5 months for patients treated surgically
(P < .01). However, neither the extent of operation
nor the achieved completeness of resection (Fig 6)
had a significant impact on survival, with a median
survival of 2.3 months for patients with gross resid-
ual disease after operation and 4 months for
patients with complete tumor resection (P = .3).
Local recurrence was seen in 20 (37%) of the 54
patients with complete or near-complete surgical
resection after a median of 2.5 months. The rate of
local recurrence did not differ between patients
with complete resection and those with minimal
residual disease (38% vs 36%; P > .6). Postoperative
XRT after complete or near-complete resection of
the primary tumor did not reduce the risk of local
recurrence (P = .15), but the time to develop that
recurrence (Fig 7) was slightly (but not significant-
ly) longer: 5 months vs 3 months; P = .08. The
median survival after a local recurrence was 66
days, and there was no difference in survival
between patients whose recurrence was treated
with XRT and those who received conservative
McIver et al 1031
Fig 6. The impact of operation on survival after a diagno-
sis of ATC. In the 96 patients in whom operation was the
primary treatment method, survival was not improved by
achieving apparently complete surgical resection. Patients
with gross residual disease postoperatively had a slightly
shorter survival (2.3 months) than those whose resection
was more complete (4 months), but these differences
were not statistically significant.
treatment (P = .44). No patient was treated surgi-
cally for locally recurrent disease.
Since 1980, 13 patients were treated with multi-
modal therapy, with debulking operation, postop-
erative XRT, and radiosensitizing chemotherapy.
The median survival among these patients does not
differ (P = .32) from the group as a whole (Fig 8),
suggesting that such a combined approach was no
more effective than operation alone in the man-
agement of these patients with ATC. Despite that, a
small number of long-term (> 1 year) survivors was
seen in this group (n = 3), a rate of 23%, compared
with less than 10% in the group as a whole.
Although not statistically significant (P = .19), this
result justifies further study in selected patients.
DISCUSSION
ATC is a rare disease, representing a diminish-
ing proportion of thyroid carcinoma presenting
to this and other institutions.10 Despite its rarity,
however, it remains one of the leading causes of
thyroid cancer mortality.2 With few cases present-
ing to any one institution, it is difficult to assess
the impact of operation and other therapy, and
there are only a few previous reports of outcomes
in large groups of ATC.3,11-16
Some tumors previously classified as ATC in fact
represent other thyroid cancer types, often with a
better prognosis. In particular, cases previously
diagnosed as small-cell ATC often stain for lym-
phocyte markers17 and are now classified as thyroid
lymphoma and should be treated accordingly.
Immunohistochemical stains for calcitonin and
1032 McIver et al
Fig 7. Time to the development of local recurrence in
patients in whom complete or near-complete resection of
the primary tumor was achieved. Postoperative XRT did
not alter the rate of local recurrence (36% vs 38%) but
slightly delayed its development (5 vs 3 months).
carcinoembryonic antigen can distinguish poorly
differentiated medullary thyroid carcinoma from
true ATC, whereas poorly differentiated (“insular”)
follicular carcinoma can be recognized by careful
histologic examination.15 It is important to rigor-
ously exclude these tumor types, which may favor-
ably bias survival rates, particularly because of the
good response to chemotherapy and radiotherapy
of some lymphomas.
This study reports the presentation and out-
comes of 134 patients with rigorously defined ATC
managed in a single institution during 5 decades.
Details of surgical and medical treatments as well as
follow-up and outcome data were available on all
patients. We confirmed the previously reported sex
distribution of ATC,13-15 with a female-to-male ratio
of 1.5:1 in this study. Similarly, the disease remains
one of middle-aged or elderly patients, with an
average age at diagnosis of 67 years. A few patients
do present at younger ages, however, and 33 of
these patients (25%) were younger than 60 years
old at diagnosis.
ATC may arise in some cases from pre-existing
differentiated thyroid cancer,1 and DTC compo-
nents are seen in a proportion of ATC cases on his-
tologic review.12,14 Current molecular models of
thyroid carcinogenesis demonstrate a stepwise
process of progressive genomic disarray, culminat-
ing in loss of the p53 tumor suppressor gene, with
anaplastic transformation.18 It remains a significant
fear for many patients and physicians that inade-
quately treated differentiated thyroid carcinoma
may undergo transformation to ATC.1 The histo-
logic review undertaken for this study revealed
DTC in 25 patients with ATC who had no previous-
Surgery
December 2001
Fig 8. Outcomes in 13 patients who received combined
treatment with debulking operation, followed by
Adriamycin chemotherapy, and XRT compared with the
remaining patients treated with “traditional” approaches
of operation or XRT. Although no difference in overall
survival is seen, a larger proportion of patients treated in
this way achieve “long-term” survival of more than 1 year
(23% vs 10%; P = .19).
ly recognized diagnosis of DTC. Only 6 patients
had a history of clinically recognized DTC.
Although anaplastic transformation of DTC can
certainly occur,19 it is a true rarity. Since most ATC
showed no evidence of a DTC component, the
stepwise model remains to be proven.
The presentation of ATC is almost always that of
a rapidly progressive enlarging neck mass, often
with symptoms of compression or invasion. Almost
half (46%) the patients had metastatic disease at
the time at diagnosis, and metastatic spread was
recognized in almost 70% of patients at some stage
in their illness. Local invasion and compression,
leading to airway compromise and aspiration pneu-
monia, was the leading cause of death.
Surgical fashions have changed throughout the
years, ranging from operation for almost no one
(1960s) to operation for all (1970s). The approach
taken during the last 2 decades has been one of
debulking or complete operation when feasible,
followed by XRT and sometimes chemotherapy.
These changes in therapeutic strategies probably
owe themselves to a persistent sense of frustration
in the management of this lethal disease.
ATC remains almost universally fatal. In this
study, mortality rates in excess of 90% were seen
within 1 year of diagnosis, and median survival was
only 3 months. These figures are worse than those
reported in several previous studies,12,20,21 though
not others,3,14,22 and this may reflect our rigorous
exclusion of other cancer types. Despite advances
in surgical technique, increasing use of XRT, and
Surgery
Volume 130, Number 6
the introduction of adjunctive radiosensitizing
chemotherapy, there is little evidence that our cur-
rent approach to this disease has yielded significant
benefit for these patients.23
Although operation was associated with a longer
life expectancy than palliation alone (3.5 months
rather than 3 weeks), part of this apparent benefit
may reflect selection bias, since patients selected for
operation are likely to have less extensive disease.
The difference in survival between patients who
underwent extensive operation and those who
underwent biopsy alone followed by XRT was small
and did not reach statistical significance (3.5 vs 2.5
months; P = .1). Once again, a selection bias toward
operation for patients with smaller tumors and less
extensive local invasion may influence these results.
Radiotherapy may also play a role in delaying
local recurrence after surgical debulking. Although
this did not reach statistical significance because of
small numbers, a useful palliation may be reflected
in this slightly prolonged time to recurrence (5 vs 3
months) seen after complete or near-complete sur-
gical resection.
Combination treatment with extensive operation,
postoperative XRT, and possibly chemotherapy
appears to carry little advantage on the basis of these
data. Even though several small pilot studies and case
reports have suggested a possible beneficial effect,6-8
the 13 patients in this study who underwent such
multimodality therapy did not demonstrate a longer
median survival. Despite this, a small number of
patients (3/13; 23%) treated this way achieved “long-
term survivor” status (> 1 year), and although a selec-
tion bias is once again possible, further clinical trials,
preferably in the form of a multicenter, randomized
prospective study, are certainly justified.
ATC is a lethal malignancy, and we have seen no
improvement in outcome during 50 years.
Operation alone is insufficient to extend life, and
more aggressive management provides no signifi-
cant additional benefit in most patients. For tumors
that are amenable to total or near-total resection,
operation followed by XRT is justified. However, the
use of adjunctive chemotherapy remains unproven
and should be submitted to clinical trials. New
approaches to the management of anaplastic thy-
roid carcinoma are desperately needed.
REFERENCES
1. Ain KB. Anaplastic thyroid carcinoma: a therapeutic chal-
lenge. Semin Surg Oncol 1999;16:64-9.
2. Ain KB. Anaplastic thyroid carcinoma: behavior, biology,
and therapeutic approaches. Thyroid 1998;8:715-26.
3. Nilsson O, Lindeberg J, Zedenius J, et al. Anaplastic giant
cell carcinoma of the thyroid gland: treatment and survival
over a 25-year period. World J Surg 1998;22:725-30.
McIver et al 1033
4. Walfish PG. Miscellaneous tumors of the thyroid. In:
Braverman L, Utiger R, editors. The thyroid. Philadelphia:
Lippincott; 1991. p. 1184-96.
5. Samaan NA, Ordonez NG. Uncommon types of thyroid can-
cer. Endocrinol Metab Clin North Am 1990;19:637-48.
6. Kobayashi T, Asakawa H, Umeshita K, et al. Treatment of 37
patients with anaplastic carcinoma of the thyroid. Head
Neck 1996;18:36-41.
7. Serrano M, Monroy C, Rodriguez-Garcia JL, et al. The com-
bined treatment of anaplastic thyroid carcinoma. Anales de
Medicina Interna 1994;11:288-90.
8. Schlumberger M, Parmentier C, Delisle MJ, et al. Com-
bination therapy for anaplastic giant cell thyroid carcino-
ma. Cancer 1991;67:564-6.
9. Hedinger CE. Histological typing of thyroid tumors. In:
Hedinger CE, editor. International histological classification
of tumors. Vol 11. Berlin: Springer-Verlag; 1988. p. 22-3.
10. Agrawal S, Rao RS, Parikh DM, et al. Histologic trends in
thyroid cancer 1969-1993: a clinico-pathologic analysis of
the relative proportion of anaplastic carcinoma of the thy-
roid. J Surg Oncol 1996;63:251-5.
11. Passler C, Scheuba C, Prager G, et al. Anaplastic (undiffer-
entiated) thyroid carcinoma (ATC). A retrospective analy-
sis. Langenbecks Arch Surg 1999;384:284-93.
12. Hadar T, Mor C, Shvero J, et al. Anaplastic carcinoma of the
thyroid. Eur J Surg Oncol 1993;19:511-6.
13. Junor EJ, Paul J, Reed NS. Anaplastic thyroid carcinoma: 91
patients treated by surgery and radiotherapy. Eur J Surg
Oncol 1992;18:83-8.
14. Venkatesh YS, Ordonez NG, Schultz PN, et al. Anaplastic
carcinoma of the thyroid. A clinicopathologic study of 121
cases. Cancer 1990;66:321-30.
15. Carcangiu ML, Steeper T, Zampi G, Rosai J. Anaplastic thy-
roid carcinoma. A study of 70 cases. Am J Clin Pathol
1985;83:135-58.
16. Nel CJ, van Heerden JA, Goellner JR, et al. Anaplastic car-
cinoma of the thyroid: a clinicopathologic study of 82 cases.
Mayo Clin Proc 1985;60:51-8.
17. Wolf BC, Sheahan K, DeCoste D, et al. Immunohisto-
chemical analysis of small cell tumors of the thyroid gland:
an Eastern Cooperative Oncology Group study. Human
Pathol 1992;23:1252-61.
18. Fagin JA. Molecular genetics of human thyroid neoplasms.
Annu Rev Med 1994;45:45-52.
19. Ozaki O, Ito K, Mimura T, Sugino K. Anaplastic transfor-
mation of papillary thyroid carcinoma in recurrent disease
in regional lymph nodes: a histologic and immunohisto-
chemical study. J Surg Oncol 1999;70:45-8.
20. Hundahl SA, Fleming ID, Fremgen AM, Menck HR. A
National Cancer Data Base report on 53,856 cases of thy-
roid carcinoma treated in the US, 1985-1995 [see com-
ments]. Cancer 1998;83:2638-48.
21. Gilliland FD, Hunt WC, Morris DM, Key CR. Prognostic fac-
tors for thyroid carcinoma. A population-based study of
15,698 cases from the Surveillance, Epidemiology and End
Results (SEER) program 1973-1991. Cancer 1997;79:564-73.
22. Lo CY, Lam KY, Wan KY. Anaplastic carcinoma of the thy-
roid. Am J Surg 1999;177:337-9.
23. Lu WT, Lin JD, Huang HS, Chao TC. Does surgery improve
the survival of patients with advanced anaplastic thyroid car-
cinoma? Otolaryngol Head Neck Surg 1998;118:728-31.
DISCUSSION
Dr Ashok Shaha (New York, NY). I enjoyed your paper.
Clearly, this is one of the largest series of anaplastic thyroid
1034 McIver et al
carcinoma from one institution. However, your experience
spans over a period of 50 years, and a retrospective study
like this is quite difficult to interpret. You have included 3
different types of anaplastic thyroid cancers, including pre-
existing thyroid disease and the presence of differentiated
carcinoma. When reviewing anaplastic thyroid carcinoma,
it is very important that we separate these groups into pre-
existing benign thyroid disease, well-differentiated carci-
noma, and the classical giant and spindle cell carcinoma. It
is also important to distinguish small cell anaplastic thyroid
cancers, which in the past were reported as anaplastic thy-
roid cancer and, on detailed immunohistochemistry, some
of them might have been lymphomas.
I was also quite surprised to see a high incidence of
surgical intervention in your series. You had 30% of the
patients having undergone complete tumor resection,
while 46% of the patients had distant metastasis at the
time of initial diagnosis. It would be interesting to see the
resection rates in various histologies. I feel that the true
incidence of surgical resectability in classical anaplastic
thyroid cancer (giant and spindle cell cancer) would be
extremely low. The incidence of 62% of patients having
surgery with curative intent is not our experience in gen-
eral. I would also appreciate it if you could comment on
these patients.
In our experience, managing the airway in patients
with anaplastic thyroid cancer has been a challenging
issue for the surgeon. Our general experience is that the
tumor does fungate through the trachea. Tracheostomy
is often necessary to maintain airway in emergent situa-
tions. The majority of anaplastic thyroid cancers we see
are extremely aggressive locally with essentially frozen
central compartments. The large mass is technically
unresectable most of the time. The small anaplastic thy-
roid cancers you have seen, measuring approximately 1.5
cm, are essentially incidental findings, and these patients
do much better in general. Again, I enjoyed your paper
and complement you on your large series.
Dr McIver. Thank you for your comments, which are
very valuable.
Regarding the airway, a large proportion of these
patients underwent tracheostomy at the time of their
primary surgery. This was almost universal when aggres-
sive surgical approaches were used, particularly in the
1970s. However, whether this is an appropriate manage-
ment strategy remains unknown. We have no evidence to
suggest that airway protection is effective, either in terms
of survival, quality of life, or quality of death. This is an
unpleasant disease to die from, and placing a tra-
cheostomy may not improve that process.
We have not subdivided our patients into those with pre-
existing thyroid disease and those without, in part because
of concern that we may have missed some prior diagnoses
of benign thyroid disease through inadequate documenta-
tion, a common problem for retrospective reviews. Only a
tiny minority was recognized to have preexisting differenti-
ated thyroid cancer, however. Although it might suggest a
different disease process from the de novo anaplastic can-
cers, they are histologically indistinguishable. It would be
Surgery
December 2001
interesting, as you suggest, to analyze that group separately
to determine if the prognosis is different, but my impres-
sion is that both types of patients fare poorly.
Like you, we were surprised to find that such a high
proportion of the patients in this retrospective review
had undergone surgery as primary treatment. I do not
think that this reflects our current experience. I believe
that the majority of patients with ATC whom we see cur-
rently, present with locally very advanced disease, and
are regarded by our surgeons as “inoperable”, for good
reason. However, some of the tumors included in this
review were small, with the smallest being a mere 1.5 cm
in maximum diameter, and surgery was pursued as pri-
mary treatment in the majority of patients who present-
ed with disease that was felt to be operable. Larger
tumors, with more local infiltration would at most be
treated with a debulking procedure, followed by exter-
nal beam irradiation.
Dr Christopher McHenry (Cleveland, Ohio). I have a
question regarding your comments about debulking.
You indicated that the median life expectancy of the
patients in your series was 2.9 months. Given the
extremely poor outcome from this disease, what is the
role for debulking? What are you trying to accomplish
with debulking?
Dr McIver. The truth is that a lot of what we have
done for this disease over the last 50 years has been dri-
ven more by a sense of frustration and a desire to do
something to help these patients, than on any basis in sci-
ence. Unfortunately, the evidence I have presented today
suggests that we are not really impacting the disease by
performing even what we would commonly regard as
“curative” surgery. Nor, indeed, have any of our other tra-
ditional modes of therapy helped.
I believe we need to be taking a much more innova-
tive and aggressive approach to chemotherapeutic regi-
mens if we are going to make any impact on this disease.
In my view, these approaches should be undertaken in
the form of multicenter trial, and I would encourage all
of us to cooperate in designing and running such a trial.
We will never get the answers otherwise.
Dr Edwin Kaplan (Chicago, Ill). I share your frustration
with this disease. We recently treated 4 patients with very
aggressive chemotherapy, including Taxol, as well as high-
dose external beam radiation therapy. Only those who did
well with this regimen were to be operated upon.
Of the 4, we ended up operating on only 1 person,
who had a complete surgical resection. She has done well
and is a long-term survivor. The other 3 developed
metastatic disease while on this very aggressive chemo-
therapy and radiation therapy. Furthermore, the
chemotherapy produced considerable morbidity.
Dr McIver. I agree entirely. In our recent experience
of treating these patients with triple therapy, it is not
uncommon to gain apparent local control over the dis-
ease in the neck, only to see metastatic disease develop.
This is rapidly progressive and results in the patients’
deaths. These outcomes add to our sense of frustration
in the face of a devastatingly lethal disease.