Abstracts / Lung Cancer 15 (1996) 381-399 391

Is follow-up of lung cancer patients after resection medically indicated and cost-effective? Walsh GL, O’Connor M, Willis KM, Milas M, Wong RS, Nesbitt JC et al. Dept. ThoracidCaniiovasadw SU~.~ M.D. Anderson Cancer Centec Box 109. 1515 Holcombe Blvd, Houston, TX 77030. Ann Thorac Surg 1995;60: 1563-70.

Background. There are no guidelines for the appropriate follow- up of patients atler pulmonary resection for lung cancer. Methods. Threo- hundred Nty-eight consecutive patients who had undergone complete resections of non-small cell lung cancer between 1987 and 1991 were evaluated for tumor recurrence and development of second primary tumors: Recurrences were categorized by site (local or distant), mode of presentation (symptomatic or asymptomatic), treatment given (curative intent or palliative), andduration ofoverall survival. Results. Recurrences developed in 135 patients (local only, 32; local and distant, 13; and distant only, 90). Of these, 102 were symptomatic and 33 were asymptomatic (most diagnosed by screening chest roentgenogram). Forty patients received treatment with curative intent (operation or radiation therapy > 50 Gy) and 95 were treated palliatively. The median survival duration from time of recurrence was 8.0 months for symptomatic patients and 16.6 months for asymptomatic patients (p = 0.008). Multivariate analysis shows that disease-free interval (greater than 12 months or less than or equal to 12 months) was the most important variable in predicting survival after recurrence and that mode of presentation, site of recurrence, initial stage, and histologic type did not significantly affect survival. New primary tumors developed in 35 patients. Conclusions. Although detection of asymptomatic recurrences gives a lead time bias of 8 to 10 months, mode of treatment and overall survival duration are not greatly affected by this earlier detection. Disease-free interval appears to be the most important determinant of survival. Screening for asymptomatic recurrences in patients who have had lung cancer is unlikely to be cost-effective. Frequent follow-up and extensive radiologic evaluation of patients after operation for lung cancer are probably unnecessary.

Elevated progastrin-releasing peptide(31-98) concentrations in pleural effusions due to small-cell lung carcinoma Shijubo N, Hirasawa M, Sasaki H, Igarashi T, Fujita A, Kodama T et al. Third Department Internal Medicine, Medico1 University School Medicine, South-l, West-16, Chuo-ku, Sapporo 060. Respiration 1996;63:106-10.

Progastrin-releasing peptide (ProGRP) is a specific and actively secreted product from small-cell lung carcinoma (SCLC) cells. Recently, an enzyme-linked immunosorbent assay, which uses monoclonal and polyclonal antibodies to recombinant ProGRP(3 l-98) a common region ofProGRP, was established. We measured concentrations ofProGRP(3 I- 98), neuron-specific enolase (NSE) and carcinoembryonic antigen (CEA) in carcinomatous and infectious pleural ellitsions. Signillcantly increased ProGRP(3 l-98) and NSE values were found in carcinomatous pleurisy due to SCLC compared to the other carcinomatous pleurisy and infectious pleurisy. CEA values were significantly increased in carcinomatous pleural effusions compared with those in infectious e5usions. The ProGRP(3 l-98) values were not correlated to NSE values in carcinomatous pleurisy due to SCLC. The determination of ProGRP(31-98) in pleural effusions will be helpful for diagnosing carcinomatous pleurisy due to SCLC.

Frequency and prognostic significance of isolated tumour cells in bone marrow of patients with non-small-cell lung cancer without overt metastases Pantel K, Izbicki J, Passlick B, Angstwurm M, Haussinger K, Thetter 0 et al. Institut Jiir Immunologic, Lumuig-Marimilians-Universitat, 80336Munchen. Lancet 1996;347:649-53.

Background. Metastasis is generally looked on as a late event in the natural history of epithelial tumours. However, the poor prognosis of patients with apparently localised lung cancer indicates that micrometastases cccur often before diagnosis of the primary tumour. Methods. At primary surgery, disseminated tmnour cells were detected immunocytochemically in bone marrow of 139 patients with non-small- cell lung carcinomas without evidence ofdistant metastases (pT,,pN,. ,Md. Tumour cells in bone-marrow aspirates were detected with monoclonal antibody CK2 against cytokeratin polypeptide 18. Patients were followed up for a median of 39 months (range 14-52) afler surgery. 2 15 patients without epithelial cancer tie, with benign epithelial tumours, nonepithelial neoplasms, or inflammatory diseases) acted as controls. Findings. In 83 of 139 (59.7%) patients cytokeratin-positive cells were detected at frequencies of 1 in 100000 to 1 in 1000000. Even without histopathological involvement of lymph nodes @NJ. tumour cells were found in 38 of 70 (54.3%) patients. 1 positive cell was found in each of 6 out of 215 controls. Surgical manipulation during primary tumour resection did not affect the frequency of these cells. In Cox’s regression analyses, the presence of such cells was a signitlcant and independent predictor for a later clinical relapse in node-negative patients @ = 0.028). Interpwtation. Early dissemination of isolated tumour cells is a frequent and intrinsic characteristic of non-small-cell lung carcinomas. The finding of these cells may help to decide whether adjuvant systemic therapy is required for the individual patient.

Transthoracic needle biopsy with a coaxially placed 2tLgauge automated cutting needle: Results in 122 patients Klein JS, Salomon G, Stewart EA. Department ofRadio/ogv. Medical Center Hosprtalof C’ermont, I11 ColchesterAve, Burlmngton, m 05401. Radiology 1996;198:715-20.

Purpose: To determine the utility of coaxial transthoracic needle biopsy (TNB) with use of a 20-gauge automated cutting biopsy needle in the diagnosis of thoracic lesions. Materials and Methods: A retrospective review was performed in 122 patients. Computed tomo- graphy was used to guide coaxial TNB, which was performed with aspirating (n = 87) and automated cutting (n = 99) needles. The sensitivities for malignant and benign lesions were determined, with a comparison of the relative yields from the two techniques. Results: The overall diagnostic yield for coaxial TNB was 88%. For malignancy the sensitivity was 95%. whereas a specific benign diagnosis was obtained in 91%. Although no difference was found for fine-needle aspiration versus core biopsy of malignant lesions (92% vs 86%) a statistically significant difference was found for benign lesions (44% vs 100%. P < .05). Pneumothorax occurred in 54%. Conclusion: Coaxial TNB performed with an automated cutting needle helps provide a diagnosis in the majority of patients with focal chest disease and is particularly useful in the diagnosis of benign lesions.

Should cavitated bronchopulmonary cancers be considered a separate entity? Mouroux I, Padovani B, Elkaim D, Richelme H. Dept. ofAbdommal/ Thoracic Surgey, Hopitol Pasteu,: B.P 69. 06002 Nice Cedex I. Ann Thorac Surg 1996;61:530-2.

Bockgmund. This study was designed to identify any clinical, histologic. and prognostic features specific to cavitated bron- chopulmonary tumors. Methods. A total of 353 patients with lung can- cer were categorized in two groups on the basis of chest radiograph and computed tomographic findings: 35 patients with cavitated cancers (group I) and 318 patients with noncavitated neoplasms (group II). Cavitation was defined as the presence of air in the tumor at the time of diagnosis and before any treatment or aspiration biopsy. The two groups were compared. ResuNs. There was no significant difference between the two groups concerning age, smoking history, or the interval to di-