ACTA OTOLARYNGOL 2001; 121; 324-328

The art of diagnosis in head and neck tumors

Alfio Ferlito, MDa,
Paolo Boccato, MDb,
Ashok R. Shaha, MD, FACSc,
Antonino Carbone, MDd,
Arnold M. Noyek, MD, FRCS(C), FACSe,
Claudio Doglioni, MDf,
Patrick J. Bradley, MBg,
Alessandra Rinaldo, MDa

a) Department of Otolaryngology-Head and Neck Surgery, University of Udine, Udine, Italy
b) Anatomic Pathology, University of Parma, Italy
c) Head and Neck Service, Memorial Sloan-Kettering Cancer Center, New York, N.Y., USA
d) Division of Pathology, Centro di Riferimento Oncologico, National Cancer Institute, Aviano, Italy
e) Department of Otolaryngology, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada
f) Division of Pathology, Belluno Hospital, Belluno, Italy
g) Department of Otolaryngology, Queens Medical Centre, Nottingham, U.K.

The diagnosis is the determination of the nature of a diseased condition, typically based on an analysis of medical history of the patient, symptoms and signs, and often on the results of laboratory tests, radiological imaging and pathological investigations. The diagnosis is the art of determining the nature of a disease.
There are many types of diagnosis: provisional or working diagnosis is an initial diagnosis and is a rational guess on which further management decisions can be predicated. As any experienced clinician knows, a working diagnosis is almost always at hand when the history alone is completed, prior to the initiation of physical examination and subsequent investigations; diagnosis by exclusion is the identification of a patient’s disease by eliminating all other known diseases; diagnosis ex juvantibus is the identification of the disease made on the basis of the patient’s response to treatment; provocative diagnosis is the induction of a condition for the purpose of a diagnosis; direct diagnosis is the diagnosis by observing pathognomonic symptoms; deductive diagnosis is the diagnosis made after the consideration of all manifestations of the disease; physical diagnosis is made on the basis of the physical examination of the patient; laboratory diagnosis is provided by the use of laboratory findings; differential diagnosis is the process of making a diagnosis considering the similarities and differences between some similar pathological conditions. The differential diagnosis includes a variety of benign and malignant conditions. Quite often the clinician will make every effort to rule out a malignant process, even though he or she may not be able to give a definite diagnosis. The classic example of this is a sore throat, where the head and neck surgeon will attempt to rule out malignancy of the laryngopharynx. There definitely are several reasons for a sore throat that may be difficult to evaluate. However, once the diagnosis of malignancy is ruled out, the patient could be treated symptomatically.
Clinical diagnosis is provided by patient’s medical history and physical examination. This is generally supplemented by imaging, endoscopy and laboratory findings.
Pathologic diagnosis is the diagnosis provided by a pathologist. The last 40 years have seen dramatic advances in the use of special studies as an adjunct to routine histological evaluation of tumors. This trend began with the advent of electron microscopy and classic cytogenetics in the early 1960s and continued with immunohistochemistry (IHC) in the late 1970s. This was advanced to DNA hybridization techniques in the 1980s, and currently with rapid molecular diagnostic protocols using polymerase chain reaction (PRC), and fluorescence in situ hybridization (FISH) in the 90s. These technical developments have increased our understanding of tumor growth control by cell cycle regulators, transcription factors, transmembrane signalling receptors, the role of oncogene and tumor suppressor gene (and apoptosis). The last ten years have seen important technological advances in the use of IHC; these advances include new methods of detection, new amplification systems, and several new methods of antigen retrieval. Very recently, automation has been introduced in the daily practice. Its main goals are to standardize diagnostic procedures and to provide reliable and reproducible results. Automation in IHC is mandatory in selected cases in which this technique provides information that may improve the process of disease management. IHC has broad applications in histopathology. For example, IHC may allow definitive and accurate classification of undifferentiated tumors, undifferentiated carcinomas and carcinomas of unknown primary site. Furthermore, IHC may solve diagnostic dilemmas in differentiating various tumors (i.e., lymphoma from carcinoma, or amelanotic melanoma from other undifferentiated tumors). IHC has proved to be of great value in differentiating neuroendocrine neoplasms of the head and neck, in particular, in distinguishing an atypical carcinoid from a paragaglioma (Ferlito et al., 1998). A number of potential prognostic factors can be readily assessed using IHC, including the proliferative growth fraction, the presence of mutations of the p53 tumor suppressor gene and the presence of p-170 glycoprotein (a product of the MDR gene). In addition, to determine treatment, microvessel quantitation, micrometastases detection in lymph nodes and bone marrow, and CD20 expression by lymphoma cells can also be studied. By IHC micrometastasis and isolated tumor cells can be detected in lymph nodes (Ferlito et al., 1999), in bone marrow (Wollenberg et al., 1994) or in other distant sites for ultrastaging system in patients with head and neck cancers (Ferlito et al., in press).
The molecular markers have made a tremendous impact on the clinical practice in reference to the ability to diagnose various small cell tumors. A small cell malignancy seen on routine histopathology under a low-power or high-power field could be subdivided into a variety of differential diagnoses, such as lymphoma, oat-cell carcinoma, amelanotic melanoma, anaplastic carcinoma, poorly-differentiated tumors, sarcoma, esthesioneuroblastoma, Kaposi's sarcoma, and eccrine carcinoma. This variety of diagnoses could be easily condensed to a single diagnosis only by using molecular markers and immunohistochemistry. There is an enormous impact of these investigations in ruling out lymphoma; the treatment of which is quite different compared to other epithelial tumors.
Cytologic diagnosis is the diagnosis provided by a cytopathologist with the analysis of individual cells, that can exfoliate spontaneously in a fluid, or collected by brushing, washing or with a fine needle (external diameter less than 0,5 mm).
For the majority of the clinicians, fine needle biopsy (with or without aspiration: FNAB, FNB) is at present considered a safe and accurate procedure for diagnosing neoplastic (benign, malignant, primary, metastatic) and non neoplastic (inflammatory, developmental anomalies) lesions in the head and neck region. In the past few decades, of the many areas of controversy in head and neck surgery, no topic has provoked as much discussion as fine needle cytology: on one hand the clinicians were still not prepared to consider its use and cytopathologists reluctant to report fine needle aspirations; on the other were those who had extensively investigated and used the technique and were convinced not only of its usefulness but also of its tremendous advantages in regard to cost effectiveness, speed of diagnosis and patient acceptability (Chen et al., 1996).
Today, for those enlightened clinicians to whom fine needle biopsy is an early part of their investigation, an interested and trained pathologist is essential and the communication between the two becomes mandatory.
There are areas of difficulty in the cytologic interpretation for which the communication between aspirator (if he/she is a clinician) and interpreter (pathologist) is always essential; but the better way is that in which the pathologist performs by him/herself the sampling, because he/she can see the patient, to gather the clinical history, to palpate the lesion, and judge using “rapid” staining(s) if the collected material is adequate for diagnostic purposes, and, sometimes, formulate a tentative or not infrequently a definitive diagnosis, in a few minutes.
In case of not palpable lesions, the sampling is usually performed by a radiologist, under echo- or TAC-guidance; in these cases too, a pathologist should always be present, in order to assure the so called “cyto-assistence” (best smearing and fixation of the collected material and check of the adequacy of the same).
The cytological diagnosis can be definitive or interlocutory; in the second case, another sampling can be planned after one month (e.g. in “reactive” lymphoadenopathy), or an histological (wedge or excisional) biopsy advised, since, in a particular case, a definitive diagnosis must be deferred to the histopathology (sometimes with the adjuvant of ancillary techniques, such as the immunohistochemistry).
The more common palpable lesions of the head and neck area observed in outpatients are mainly related to the thyroid, lymph nodes, major salivary glands (Boccato et al., 1998): in a high percentage of these cases, the cytological diagnosis performed on smeared material collected with a fine needle allow a definitive diagnosis; the more frequent exception to that rule are the so called “follicular neoplasia“ of the thyroid, some rare tumors of the parotid, a frequent, often deceptive salivary neoplasm such as the mucoepidermoid carcinoma, and the primary lymphoproliferative diseases. In the last cases a fine needle cytodiagnosis is per se alone considered insufficient, and the diagnostic confirmation of a lymphomatous lesions (and the subclassification of it) is deferred to the histopathology.
As far as the lymph nodes are concerned, FNAB and FNB can avoid unnecessary surgery in benign (“reactive”) cases, if adequately followed; the same can be true for the frequent Warthin’ tumor in subjects at high surgical risk, and in many other cases such as nodular goiter, thyroiditis, metastatic lesions of a known primary.
The importance of FNAB and FNB in establishing a diagnosis in a high percentage of cases with the minimal intervention is well appreciated and there is an argument to be made for creating one-step fine needle sampling in the clinics (as e.g. for the breast surgery) for the (rapid) diagnosis of head and neck palpable and not palpable lesions (Weir and Shah, 2000). It is extremely important for the clinician to understand the pitfalls of fine needle aspiration biopsy. As a matter of fact, it is advisable to tell the patient well in advance that a fine-needle aspiration biopsy is only one more investigation and not a final diagnosis. Unfortunately, many times the patient tends to rely heavily on the results of the FNA biopsy to rule in or rule out malignancy. If the FNA biopsy does not conform to the clinical diagnosis, more weight should be given to the clinical diagnosis and further evaluation should be undertaken. Even though the results of the core biopsy and the FNA biopsy are equivalent, occasionally a core biopsy may be of definite benefit. In a large group of lymph nodes where a diagnosis of lymphoma may be suspected, a core biopsy will give adequate tissue for appropriate tests. Additionally, in patients with anaplastic carcinoma where surgery is not indicated, a core biopsy will document the presence of giant and spindle carcinoma. Fine-needle aspiration biopsy has made a tremendous impact on the clinical practice of head and neck tumors and is probably the most appropriate initial test in the evaluation of patients presenting with head and neck pathologies. Clearly, this is the most cost-effective test available today in head and neck tumor diagnosis. It is extremely important for the clinician to communicate directly with the cytopathologist and discuss the clinical findings, in light of which the fine-needle aspiration biopsy will be of enormous diagnostic accuracy. Pitfalls of needle biopsy aim care biopsy.
Frozen-section diagnosis. The diagnosis is made examining a frozen section of tissue during the course of a surgical procedure. In particular, when the surgeon requires guidance before proceeding further, with reference to the infiltrative or suspect aspects of the resection margins. Frozen section assessment of the excision margins is a procedure planned in most protocols used in head and neck tumor surgery. The issue of frozen section continues to generate considerable controversy, among both the clinicians and the pathologists. The frozen section is frequently indicated and the results will make a difference in the intraoperative decision-making or the extent of surgery. However, there are several factors involved in frozen section. The frozen section diagnosis is very easy in conditions such as squamous cell carcinoma, but it may be impossible to make a diagnosis of lymphoma under frozen section. It is also extremely useful for margins of resection of tumor in epithelial malignancies. However, its role in the diagnosis of follicular neoplasms and certain salivary tumors is controversial. If the surgeon is going to undertake a major resection based on the frozen section, it is important to make certain it conforms to the clinical diagnosis. A second opinion may be obtained from another pathologist.
Final or permanent section diagnosis. The pathologist often uses the term “final” diagnosis to indicate the diagnosis made on permanent sections in contrast with frozen-section diagnosis. Also a cytologic diagnosis may be definitive (e.g. lymph node metastasis of well differentiated squamous cell carcinoma, papillary carcinoma of thyroid, others). Morevover some cryostatic aspects can be per se insufficient for a sure definitive diagnosis , and that must be deferred to definitive preparations , and/or to IHC.
Molecular diagnosis. Applications of molecular genetic techniques for the diagnosis, prognosis and treatment to clinical samples is a new and promising field: these methods, probably, will complement and eventually replace in the near future other, more subjective diagnostic techniques. Furthermore they will become necessary in delineating the genetic profile of any single case in order to provide the guidelines for tailoring gene therapy. However, at present, their role in daily practice is quite limited. They represent a formidable research tool for the discovery of genes involved in head and neck cancerogenesis, identification of molecular alterations associated with prognosis and chemo/radioresistence, but translation of this new information in the clinical practice is still in its infancy.
Qualitative diagnosis. The diagnosis made by pathologist is a qualitative diagnosis. Quantitative diagnosis. The quantitative diagnosis is the identification of a quantitative lesion utilizing radiologic imaging, therefore it is provided by radiologist. Level diagnosis or niveau diagnosis is the localization of the exact level of the lesion.
Imaging and pathological methods are often necessary to make a correct diagnosis. The diagnosis is usually provided by pathologist and suggested by radiologist. Treatment decisions in patients with neoplasms should always be based on pathological diagnosis. An incorrect diagnosis of a tumor may result in severe malfunction and even loss of the organ. In the past laryngectomy was performed without previous biopsy and pathology museum contain many examples of a wrong clinical diagnosis (Friedmann ad Ferlito, 1988). The pathologist usually provides a qualitative diagnosis indicating that the lesion is a squamous cell carcinoma or another oncotype. The radiologist often provides a quantitative diagnosis and may suggest a specific and differential diagnosis. For instance the radiologist often is in the position to establish if the lesion is “cystic” or “infiltrative” or “vascular”. The radiologist may indicate cartilaginous neoplastic lesions but it is impossible to distinguish benign from low-grade malignant chondroid lesions based on imaging characteristics unless there are aggressive features. The revolution in diagnostic imaging in recent years has had a major impact on the practice of medicine. The role of the radiologist and pathologist are complementary to the practing clinician.

The art of diagnosis reflects the knowledge, competence and practice of a team of physicians, often faceless and nameless (Noyek, 1978).
To avoid mismanagement of the patient and complications if is mandatory that the correct diagnosis is established prior to the definitive treatment. The choice of treatment is an important parameter influencing not only prognosis but also, more importantly, the patient’s quality of life. Obviously, prognosis is better if the most appropriate therapy is undertaken initially. Apparent discrepancies between pathological and clinical findings should be considered carefully and may indicate the need for further pathological, radiological and clinical investigations. Before any treatment we have to formulate an accurate diagnosis. Regardless the diagnostic method, our goal is to make an early and in particular accurate diagnosis and then we can promote an adequate therapy.
The art of diagnosis fully reflects human element. The acquisition of individual bits of scientific evidence is purely objective; however the art of collating and manipulating this information, and subsequently arriving at a correct diagnosis, is largely subjective. Pieces of diagnostic information may be fed into a computer, in the hope of arriving at a single valid diagnosis, but it is a human who must program this computer, and therein lies the rub (Noyek, 1978). The issues related to the correct diagnosis and the available tools in making an appropriate diagnosis remain controversial. However, the roles of fine-needle aspiration biopsy, immunohistochemistry, and molecular markers have made a major impact in the recent past. It is also extremely important for the clinician to have close communication with the radiologist and pathologist. Occasionally, in a given patient, the diagnosis may be very difficult and one may consider a diagnosis made by exclusion. A follow-up may be necessary to rule out a malignant process and this is especially true in the evaluation of suspected recurrent tumors. The diagnosis and evaluation of recurrent tumors continues to be a challenge to the clinician, since it is almost impossible to rule out recurrent tumor in view of the post-surgical or post-radiation therapy changes. Under these circumstances, a close clinical follow-up may be more important to rule out recurrent malignant tumor. In rare instances, the final diagnosis may be made at the time of autopsy, should it be undertaken. Even though the incidence of autopsy has gone down considerably in recent years, the value of an autopsy continues to be significant (Baron, 2000). The diagnosis, differential diagnosis, final diagnosis, and making a correct diagnosis continue to be an evolving process in the lifetime of the physician and the patient's disease condition.
Noyek (1978) pointed out that “Accurate diagnosis is the only true cornerstone on which rational treatment can be built; and once we have arrived at an accurate diagnosis, our universal aim should be to treat every patient as we would wish our families or, indeed, ourselves to be treated, acknowledging all the pressures that society brings to bear upon us as physicians, at this time”.
In terms of working practice, the need of knowledge, communication, transparency and competency is of paramount importance in making an accurate diagnosis, but as humans, we remain fallible. Hippocrates knew this: “I warmly commend the physician who makes small mistakes: infallibility is rarely to be seen” (Baron, 2000).

References

Baron JH. Clinical diagnosis and the function of necropsy. J R Soc Med 2000; 93: 463-466.

Boccato P, Altavilla G, Blandamura S. Fine needle aspiration biopsy of salivary gland lesions. A reappraisal of pitfalls and problems Acta Cytol 1998; 32: 888-898.

Chen VSM, Qizilbash A, Young JEM. Head and Neck. In: Guides to Clinical Aspiration Biopsy, ed 2. New York, Tokio: Jgaku-Shoin, 1996.

Ferlito A, Devaney KO, Rinaldo A. Is it time to develop an “ultrastaging system” for use in patients with head and neck malignancies? Laryngoscope. In press. (Commentary).

Ferlito A, Barnes L, Rinaldo A, Gnepp DR, Milroy CM. A review of neuroendocrine neoplasms of the larynx: update on diagnosis and treatment. J Laryngol Otol 1998; 112: 827-834.

Ferlito A, Devaney KO, Rinaldo A, Devaney SL, Carbone A. Micrometastases: have they an impact on prognosis? Ann Otol Rhinol Laryngol 1999; 108: 1185-1189.

Friedmann I, Ferlito A. Granulomas and Neoplasms of the Larynx. Edinburgh: Churchill Livingstone, 1988: p. 129.

Noyek AM. Some comments on the art of diagnosis. Otolaryngol Clin North Am 1978; 2: 247-249.

Weir J, Shah K. The future of ORL-HNS and associated specialties series. The future of pathology. J Laryngol Otol 2000; 114: 491-493.

Wollenberg B, Ollesch A, Maag K, Funke I, Wilmes E. Mikrometastases im Knochenmark von Patienten mit Karzinomen des Kopf-Hals-Bereiches. Laryngorhinootologie 1994; 73: 88-93.