POORLY DIFFERENTIATED THYROID
CARCINOMA: INTRODUCTION TO THE
ISSUE, ITS LANDMARKS, AND
CLINICAL IMPACT
Juan Rosai, M.D.
From the Department of Pathology,
National Cancer Institute,
Milan, Italy
1
In his book on Human Tumors, Pierre Masson lamented that "No classification is more
difficult to establish than that of thyroid [carcinomas]. Their pleomorphism is almost the rule; very
few are adapted to a precise classification" [1]. There are few issues that support this statement as
strongly as those concerning the existence and definition of poorly differentiated thyroid carcinoma
(PDTC). Since the idea was first proposed, some authors have enthusiastically endorsed it, others
have grudgingly accepted it after some initial hesitation, and still others remain profoundly skeptical
[2-6]. Since it would be ludicrous to think that these various positions are in any way influenced by
other than scientific considerations, it is only those that we will address in this brief contribution.
Let's start by saying that nobody who has seen a typical case of insular thyroid carcinoma will deny
that the tumor has a distinctive appearance, almost to the point of instant pattern recognition, and
that it would be inaccurate and misleading to designate such a tumor either as a follicular or
papillary carcinoma without a qualifier, or as an undifferentiated (anaplastic) carcinoma. Alas, to
provide a satisfactory definition of a poorly differentiated carcinoma has proved an almost
insurmountable task. It reminds one of the famous crack about pornography to the effect that "I
don't know how to define it, but I have no proplems in recognizing it when I see it".
Much of the trouble concerning the identify of PDTC can be traced back to the original
proposals. Independently and almost simultaneously, Carcangiu et al [7] and Sakamoto et al [8]
proposed the concept using a similar terminology but widely disparate criteria. Carcangiu et al [7]
described a tumor with a distinctive set of architectural and cytologic features, which included
insular pattern of growth (hence the name), small cell size, round hyperchromatic nuclei, mitotic
activity, and frequent necrosis. Sakamoto et al [8] based instead their proposal exclusively on the
pattern of growth: tumors with a follicular or papillary pattern were regarded as well-differentiated
(although the latter assumption can be challenged, since the normal thyroid does not make papillae),
whereas tumors with a solid, trabecular, or sclerotic ("scirrhous") pattern of growth were
categorized as poorly differentiated. It should obvious that the first proposal is an attempt at the
description of a distinct "entity", whereas the latter is a grouping of a morphologically diverse group
2
of tumors on the basis of architectural similarities. In this context, the fact that Carcangiu et al [7]
entitled their paper poorly differentiated ("insular") carcinoma was not particularly fortunate, since
this could be taken to imply that the two terms are synonymous, when what was meant was that
insular carcinoma was a type (one may say the prototypical expression) of a poorly differentiated
carcinoma composed of follicular cells. Unfortunately, this dualistic approach has not only persisted
but become accentuated in recent years. Whereas some authors restrict the term PDTC to tumors
that fulfill to the original description of insular carcinoma (whether using that designation or others,
such as primordial cell carcinoma) [9-11], others include not only insular carcinomas, but also any
type of carcinoma with a predominantly solid/trabecular pattern of growth, as well as tumors that
have been found to be more aggressive than the usual papillary or follicular carcinoma even if their
morphology is, strickly speaking, not poorly differentiated (such as columnar cell carcinoma or
mucoepidermoid carcinoma) [5, 12-15]. The quandary is obvious.
Another criticism that has been raised is that the use of the term insular carcinoma implies
the existence of a distinct tumor type for what may be instead a lesser differentiated form of one or
another of the two major types of well-differentiated thyroid carcinoma, i.e., papillary and follicular
[5]. According to this view, there is not a single PDTC (not even if one restricts the term to the
prototypical insular carcinoma) but rather poorly differentiated forms of either papillary or follicular
carcinoma. It should be noted that this position is not too different from that espoused in the article
by Carcangiu et al [7], as graphically depicted by the diagram shown in Fig.1, which in its utter
simplicity tells this story: (1) thyroid tumors of follicular cells show a wide range of differentiation,
which tends to correlate with their degree of aggressiveness; (2) the two major types are papillary
and follicular carcinoma, a distintion originally defined on morphologic grounds and now supported
by molecular markers; (3) the differences among those two major types are obvious among the
well-differentiated forms, but become increasingly blurred as the tumors become less differentiated,
to disappear altogether in the undifferentiated forms. The corollary of this scheme is that there may
3
be well be two kinds of poorly differentiated carcinomas, but that the very fact that they are poorly
differentiated matters much more than to the subtype they may belong.
What practical messages emerge from these considerations, where one to accept their
premises ?
The first is that insular carcinoma is not to be equated with PDTC but rather regarded as one
of its subsets.
The second is that both typing and grading play an important role in the evaluation of
thyroid tumors (as they do in several other classification systems, such as breast carcinoma and soft
tissue sarcoma), and that there is nothing intrinsically wrong with combining them in a
classification scheme on the basis of their relative clinical significance.
The third and most important message – as usual – is the realization that much still needs to
be done. As above stated, the original description of insular and other types of PDTCs included a
long list of morphologic features, both of architectural and cytologic type. When all of them are
present in a given tumor, an aggressive behavior can be anticipated. But what if one or another of
those features are missing, as is often the case? Are some of them prognostically more significant
than others? Let's illustrate the point with some concrete examples:
1) Is a tumor with an insular pattern of growth but the nuclear features of papillary carcinoma
more or less aggressive than a tumor showing a papillary configuration but composed of
small, hyperchromatic, mitotically active nuclei?
2) Is it worse to have an architecturally and cytologically typical insular carcinoma that shows
minimal or no invasion or a better differentiated tumor with widespread invasion?
3) What is the clinical significance of the presence of focal poorly differentiated features in a
predominantly well-differentiated thyroid carcinoma? [16, 17]
4) What role will special techniques (read: molecular) play in this determination?
In conclusion, it seems to me that the concept of PDTC has stood the test of time and proven its
validity, at least on general grounds. For whatever it is worth, it has received the imprimatur of the
4
latest WHO Committee on Thyroid Tumors and has been incorporated into their classificaton
scheme [18]. However, its definition remains blurred. It will be the task of the current generation of
thyroid pathologists to put it in focus.
PAPILLARY
CARCINOMA
POORLY
DIFFERENTIATED
CARCINOMA
UNDIFFERENTIATED
CARCINOMA
FOLLICULAR
CARCINOMA
FIGURE 1
Position of poorly differentiated carcinoma in the scheme of thyroid carcinoma composed of
follicular cells.
5
REFERENCES
1. Masson P. Human tumors. Histology, diagnosis and technique, 2nd Edition, Detroit, Wayne
State University Press, p.589, 1970.
2. Akslen LA, LiVolsi V. Poorly differentiated thyroid carcinoma – it is important (letter to the
editor). Am J Surg Pathol 24: 310, 2000.
3. Mizukami Y, Nonomura A, Michigishi T, Noguchi M, Ohmura K, Nakamura S, Hoso M.
Poorly differentiated (“insular”) cacinoma of the thyroid. Pathol Int 45: 663-668, 1995.
4. Nishiyama RH. Another dissertation on poorly differentiated carcinoma. Is it really
necessary? Adv Anat Pathol 6: 281-286, 1999.
5. Pilotti S, Collini P, Mariani L, Piacucci M, Bongarzone I, Vigneri P, Cipriani S, Falcetta F,
Miceli R, Pierotti MA, Rilke F. Insular carcinoma: A distinct de novo entity among
follicular carcinomas of the thyroid gland. Am J Surg Pathol 21: 1466-1473, 1997.
6. Sobrinho-Simoes M. Poorly differentiated carcinomas of the thyroid. Endocr Pathol 7: 99102, 1996.
7. Carcangiu ML, Zampi G, Rosai J. Poorly differentiated (“insular”) thyroid carcinoma: A
reinterpretation of Langhans’ “Wuchernde struma”. Am J Surg Pathol 8: 655-668, 1984.
8. Sakamoto A, Kasai N, Sugano H. Poorly differentiated carcinoma of the thyroid. A
clinicopathologic entity for a high-risk group of papillary and follicular carcinomas. Cancer
52: 1849-1855, 1983.
9. Decaussin M, Bernard MH, Adeleine P, Treilleux I, Peix JL, Pugest M, Tourniaire J, Berger
N. Thyroid carcinomas with distant metastases: A review of 111 cases with emphasis on the
prognostic significance of an insular component. Am J Surg Pathol 26: 1007-1015, 2002.
6
10. Papotti M, Botto Micca F, Favero A, Palestini N, Bussolati G. Poorly differentiated thyroid
carcinomas with primordial cell component: A group of aggressive lesions sharing insular,
trabecular, and solid pattern. Am J Surg Pathol 17: 291-301, 1993.
11. Pellegriti G, Giuffrida D, Scollo C, Vigneri R, Regalbuto C, Squatrito S, Belfiore A. Cancer
95: 2076-2085, 2002.
12. Mizumani Y, Nakajima H, Annen Y, Michigishi T, Nonomura A, Nakamura S. Mucinproducing poorly differentiated adenocarcinoma of the thyroid. A case report. Pathol Res
Pract 189: 608-611, 1993.
13. Papotti M, Torchio B, Grassi L, Favero A, Bussolati G. Poorly differentiated oxyphilic
(Hurthle cell) carcinomas of the thyroid. Am J Surg Pathol 20: 686-694, 1996.
14. Sobrinho-Simoes M, Nesland JM, Johannessen JV. Columnar-cell carcinoma; another
variant of poorly differentiated carcinoma of the thyroid. Am J Clin Pathol 89: 264-267,
1988-89.
15. Sobrinho-Simoes M, Sambade C, Fonseca E, Soares P. Poorly differentiated carcinomas of
the thyroid gland: A review of the clinicopathologic features of a series of 28 cases of a
heterogeneous, clinically aggressive group of thyroid tumors. Int J Surg Pathol 10: 123-132,
2002.
16. Ashfag R, Vutch F, Delgado R, Albores-Saavedra J. Papillary and follicular thyroid
carcinomas with an insular component. Cancer 73: 416-423, 1994.
17. Sasaki A, Daa T, Kashima K, Yokoyama S, Nakayama I, Noguchi S. Insular component as a
risk factor of thyroid carcinoma. Pathol Int 46: 939-946, 1995.
18. Tumors of endocrine organs. Pathology and Genetics, World Health Organization
Classification of Tumours, Lyon, 2004 (in press).
7