Idiopathic pneumonia syndrome after high-dose chemotherapy for relapsed Hodgkin's disease

• Published in: British journal of cancer 1997, Vol.75 (7), p.1044-1048
• Main Authors: Rubio, C, Hill, ME, Milan, S, O'Brien, MER, Cunningham, D
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Antineoplastic agents; Antineoplastic Agents, Alkylating - adverse effects; Biological and medical sciences; Biomedical and Life Sciences; Biomedicine; Cancer Research; Carmustine - adverse effects; Chemotherapy; clinical-oncology; Drug Resistance; Epidemiology; Female; Hodgkin Disease - complications; Humans; Male; Medical sciences; Middle Aged; Molecular Medicine; Multivariate Analysis; Oncology; Pharmacology. Drug treatments; Pneumonia - etiology; Proportional Hazards Models; Risk Factors; Sex Factors; Syndrome
• ISSN: 0007-0920; 1532-1827
• DOI: 10.1038/bjc.1997.178

The risk of idiopathic pneumonia syndrome (IPS) in patients with Hodgkin's disease (HD) undergoing high-dose chemotherapy (HDC) is significant, and once developed IPS is potentially fatal. The aim of this study was to quantify this risk accurately and determine prognostic factors for its development and course. Using a computerized database, all patients with HD treated with BCNU (carmustine) containing HDC and haematopoietic support at The Royal Marsden between November 1985 and March 1994 were identified. Patient characteristics, previous treatments, disease status at HDC, dose of BCNU, incidence and severity of IPS and survival were all determined and analysed. During the study period, 94 patients received HDC, of whom 26 (28%) had a first episode of IPS within a year of HDC and 23 within 6 months. The median time to presentation after HDC was 93 days (range 12-336 days). The only factors that significantly increased the risk of developing IPS on multivariate analysis were dose of BCNU (P for trend = 0.03) and female sex (P = 0.04). Of these 26 patients, 14 had complete resolution of all symptoms, three had persisting pulmonary symptoms at 6 months and the remaining nine died of IPS at a median of 74 days (19-418 days). All the patients who died from IPS had the first symptoms within 6 months of HDC and all received doses of BCNU > 475 mg m(-2) (P for trend = 0.001). For women receiving > 475 mg m(-2) the risk of death was significantly higher than for men (P = 0.035) but not for those receiving < 475 mg m(-2). Previous lung disease, persisting residual disease before HDC, previous bleomycin or previous mantle radiotherapy did not increase either the incidence of IPS or risk of a fatal outcome. We conclude that the main avoidable risk factor for fatal IPS after HDC is dose of BCNU, and this is especially true for women. If < 475 mg m(-2) is given, even patients with previous mantle radiotherapy and/or previous bleomycin have a very low risk of developing fatal lung toxicity if lung function tests are normal.