Second primary tumours occur frequently in patients with a history of head and neck malignancies. Delays in making an early and correct diagnosis can seriously affect the therapy management and survival.

This was a retrospective study of 120 patients with a history of head and neck cancer, presenting with a second primary tumour. Current follow-up strategies and the use of routine sonographic imaging of the head and neck regions were evaluated, and the impact that tumour chronology, the tumour site and the various treatment modalities have on the survival were assessed. Forty-two per cent of patients developed a metachronous second malignancy more than five years after diagnosis of the index tumour. The accuracy of colour-duplex sonography in detection of second primaries in the head and neck was 82.3 per cent. First and second primary tumours located in the larynx were observed to have the highest five-year survival rate. Patients who developed metachronous tumours had a five-year survival rate of 68.9 per cent for the index tumours, and a 26 per cent five-year survival rate with the occurrence of a second neoplasm. With synchronous tumours a mean survival time of 18 months and a five-year survival rate of 11.9 per cent was found (p < 0.0001). Where clinically appropriate an aggressive treatment strategy was employed and yielded the most favourable results with a five-year survival rate of 66.8 per cent and 35.9 per cent for index tumours and second primary malignancies, respectively.

Since more than 40 per cent of the metachronous second primaries in patients with a history of head and neck malignancy occur beyond the five-year follow-up period, an extended protocol with individually adjusted close monitoring of high-risk patients seems appropriate. Colour-duplex sonography is a valuable screening investigation for the early detection of second primary tumours. The treatment of a second primary is often less successful than for the same malignancy occurring primarily. The prognosis of synchronous tumours is significantly lower when compared to malignancies of a metachronous nature, despite some encouraging individual results. Only the early implementation of aggressive treatment methods for second primaries is successful in terms of survival.

A retrospective review of 240 patients with T1/T2 squamous cell carcinomas of the larynx was performed. Seventy-two per cent had glottic primaries, 27 per cent had supraglottic tumours and one per cent had subglottic disease. Sixty-nine per cent presented with T1 disease and 31 per cent had T2 staged tumours. All patients were treated with definitive radiotherapy between 1973 and 1997.

With a median follow-up of 68 months, 68 patients (28 per cent) have developed 72 other cancers. Ten of 68 presented with synchronous primaries (15 per cent). Thirty per cent of glottic patients and 25 per cent of the supraglottic/subglottic patients developed second cancers. The most frequent second malignancy was lung cancer: 28/72 (39 per cent). Fifteen patients developed second head and neck cancers (21 per cent). Other second primary sites included oesophagus (eight), prostate (six), colorectal (five), breast (two) and others (eight). The median time from radiotherapy until the development of a second cancer was 31 months. The Kaplan-Meier survival estimate at five years was significantly less for those patients developing second cancers (55 per cent) compared to those not developing second malignancies (70 per cent), (p<0.05). The median survival from the development of a second cancer was 14 months. More died as a result of a second cancer (41 patients) than their primary laryngeal cancer (40 patients). Second cancers are common and deadly in patients with early stage laryngeal carcinoma.

The phenomenon of multicentricity in head and neck squamous cell carcinoma affects survival rates. We evaluated the use of panendoscopy/triple endoscopy and, in particular, the place of bronchial washings in the initial staging assessment of head and neck tumours. In a prospective panendoscopic study, a second primary rate of 4.8 per cent was discovered. All four bronchial tumours – both index and simultaneous primary cases – were obvious on bronchoscopy and chest X-ray, despite 50 per cent of them being clinically silent. A self-limiting complication rate of 2.4 per cent was encountered. The issue of bronchial washings was debated and our sensitivity (50 per cent) and specificity (97 per cent) results assessed. We advocate the inclusion of bronchoscopy as part of a panendoscopic work-up of head and neck tumours. The rationale for this is discussed. Bronchial washings, although cheap and easy to process, did not change the management of any patient in the study group and contamination could complicate assessment. Overall, panendoscopy is a safe worthwhile procedure.

Abnormalities of p53 tumour suppressor gene are detected in a diversity of malignancies and play an important role in their pathogenesis. Hypopharyngeal carcinoma is the most morbid among head and neck squamous cell carcinomas because of the high incidence of treatment failures and because a biological marker predictive of the treatment failures remains elusive. The expression of p53 protein in 46 hypopharyngeal squamous cell carcinomas was examined histochemically and p53 immunoreactivity was found in 19 of 46 cases (41.3 per cent). The rate of second primary tumour development was significantly higher in the p53-positive group than in the p53-negative group (p = 0.039), whereas that of tumour recurrence was not significantly different between the two. Moreover, there was no statistically significant difference in either overall or disease-free survival between the p53-positive and -negative groups. These results indicate that although p53 expression significantly correlates with second primary tumour development in patients with hypopharyngeal squamous cell carcinomas, it is not predictive of the clinical outcome.

Synchronous or metachronous second primary malignancies of the lung are sometimes encountered in patients with laryngeal cancer while the occurrence of a laryngeal second primary following cancer of the lung is rare.

A two-armed study was conducted. A prospective arm in which the larynges of 56 terminal lung cancer patients were examined, and a retrospective arm incorporating both a chart study of 126 terminal head and neck cancer patients (HNCP) and a computerized search of all hospital records of patients with laryngeal and lung cancers. No laryngeal malignancy was found in the lung cancer patients' group and no antedating pulmonary malignancy was recorded in the terminal HNCP. The computerized search of 1778 lung cancer patients and 213 laryngeal cancer patients also failed to demonstrate cases where the former preceded the latter.

In conclusion. No second primary of the larynx was found in lung cancer patients. These results compare with reports of large databases where cancer of the larynx was found in a negligible percentage of lung cancer survivors and theories explaining this are discussed.