Release: July 29, 2000
Contact: Kenneth Satterfield
703-519-1563
[email protected]
In San Francisco (7/28-8/2)
415-284-8082
A DELAY IN DIAGNOSIS DOES NOT IMPACT ON EXTENT OF HEAD AND NECK CANCER
San Francisco, CA -- Despite public policy initiatives to increase cancer awareness, patients are delaying diagnosis of some head and neck cancers. Furthermore, a new study suggests the delay has not proven to have an impact on the extent of disease when found. If these findings are confirmed and found to extend to other cancer types, as is likely, then it will have important ramification on future health strategy for cancer in general and specifically head and neck.
The authors of the study, "Delay in Diagnosis and its Effect on Outcome in Head and Neck Cancer" are Mark McGurk, MD FRCS, Chi-Hwa Chan FDS, Esther O'Regen BDS, and Gitta Madani MBBS, all from Guy's Hospital, London, United Kingdom. Their findings were presented before the 5th International Conference on Head and Neck Cancer, being held July 29 through August 2, at the San Francisco Marriott, San Francisco, CA. More than 1,500 leading head and neck surgeons from the United States and 46 nations will gather to hear the latest medical research in the diagnosis, treatment, and reconstruction associated with head and neck cancer. The medical conference is sponsored by the American Head and Neck Society, www.headandneckcancer.org.
Methodology: The study data were based on two cohorts of patients with benign and malignant tumors of the head and neck, principally squamous cell carcinoma of the mouth, pharynx and larynx but also thyroid, salivary gland and skin tumors.
The first cohort was a retrospective series of 930 consecutive patients treated between 1961 and 1986; data on delay in presentation was available on 695 of these patients of whom 394 had upper aerodigestive cancer. The second cohort consisted of a prospective series of 368 consecutive patients treated from 1992-1999. In that group data on delay was available in 332 patients of whom 205 had upper aerodigestive cancer. The patient characteristics of those with squamous cell cancer were similar in the two cohorts, the main difference being the first cohort had a greater proportion of laryngeal and thyroid neoplasms and the latter cohort, more mouth cancers. Laryngeal cancer tends to present at an earlier stage of the disease and has a good outcome.
Delay was arbitrary defined as three months or longer from the onset of symptoms to treatment. The surgeon elicited the period of delay in the three-week period between diagnosis and commencement of treatment. In the prospective series the patient was questioned on the circumstances of delay on two separate occasions during this period. Responsibility for delay was apportioned to the patient or physician if the period of postponement was greater than 6 weeks at any point in the referral pathway (patient to doctor; doctor to hospital, hospital to treatment).
If more than one person was involved then responsibility was attributed to the individual responsible for the longer period of delay. Staging in this study was clinically based and simplified into early (stage I or II) and advanced (stage III or IV). For benign tumors, T size was considered equivalent to stage.
Details were available in both cohorts on a range of both patient and tumor factors; age, gender, race, social status, performance status (ECOG 1-5), history of a previous cancer, tumor site, histological grade and tumor type (benign, or malignant) although details on race, marital status, alcohol and cigarette consumption were available only in the prospective data set.
The Fischer's exact test was used to test for differences in variables between the two study groups. The endpoints used for the outcome of delay, in patients with upper aerodigestive cancer, were tumor stage and disease specific survival. The relationship between delay, tumor stage at presentation and survival were tested by multiple regression analysis and survival curves generated by the Caplan Mier method. The influence of patient variable on predicting delay or extent of disease at diagnosis was tested in a multiple linear regression model, and the extent of disease at diagnosis was investigated by logistic regression.
Results: Major findings included:
The mean age of patients with upper aerodigestive cancer was 61 years in the retrospective and 62 years in the prospective cohort of patients. The male to female ratio was 1.6:1. The median period of delay in each cohort was three months. The proportion of patients with delay (retrospective series =50.3 percent and prospective series =56.6 percent) and advanced disease (retrospective =46.7 percent and prospective = 54.5 percent) was similar in each cohort and remained unchanged over 40 consecutive years. But the patients with delay were not the same as those with advanced disease, and there was no correlation between delay and either stage (p=0.8 & p=0.8) at presentation or subsequent survival (p=0.63 and p=0.8) in both study groups.
When the analysis was repeated with delay as the dependant variable, only benign disease (retrospective p=0.0001: prospective p=0.004) predicted for delay in presentation. A similar result was observed with the indolent cancers of the skin (p=00001) and salivary gland (retrospective p=0.0001: prospective p=0.06) in a logistic regression analysis.
The first cohort consisted of patients from a district general hospital; the second, from a teaching hospital with a line of secondary referral from a dental school. With head and neck neoplasms and upper aerodigestive cancers the persons mainly responsibility for delay in diagnosis were the patients themselves. Although the median delay for aerodigestive cancers was three months, it varied little by site in the upper respiratory /alimentary tract (oral cavity: pharynx: larynx) but in a considerable degree with skin (eight months), salivary (18 months), glands and thyroid neoplasms (two months).
The median delay in diagnosis for aerodigestive cancer was three months from symptom recognition to treatment, but it was 10.5 months for benign tumors of salivary, skin and thyroid cancers. The likely explanation is that rapidly growing lesions with overt symptoms such as pain or bleeding are frightening. In both cohorts, the patient was responsible for most of the delay (78 percent and 80.6 percent, respectively). The delay due to physicians was equally distributed between general practitioners and the specialists; risk of hospital delay was increased if the patient was referred to a general rather than a cancer center.
Conclusions: Half the patients with upper aerodigestive cancers present with advanced disease, and the incidence has not changed over the past four decades. In these cancers, delay in diagnosis does not correlate with stage of tumor at presentation nor survival. The most likely explanation is that many tumors are relatively silent or symptoms go unrecognized until they are advanced in development.
- end -
For additional information regarding head and neck cancer, log on to http://www.entnet.org/cancer.html
RSS