Risk factors of treatment-related death in chemotherapy and thoracic radiotherapy for lung cancer

Abstract

We retrospectively analysed the incidence and risk factors of treatment-related death in the treatment of chemotherapy- and thoracic radiotherapy-naı̈ve patients with lung cancer. Between July 1992 and December 1997, 1799 patients were diagnosed as having lung cancer in our hospital and 926 patients received chemotherapy and/or thoracic radiotherapy. 25 patients (2.7%) died from toxicity of the treatment, 10 from pneumonia, 7 from radiation pneumonitis, 6 from sepsis, 1 from perforation of the small intestine and 1 for an unknown reason. 18 patients (2.3%) died from chemotherapy-related toxicity. The incidence of treatment-related death (TRD) from chemotherapy was highly correlated with the performance status (PS), PS 0: 0.7%, PS 1: 2.2%, PS 2: 4.0%, PS 3: 7.7% and PS 4: 25% (P=0.004). 7 patients (1.6%) died from pneumonitis after thoracic radiotherapy. Multivariate analyses demonstrated that poor PS (relative risk (RR): 1.95, 95% confidence interval (CI): 1.05–3.65, P=0.034) and chemotherapy using the cisplatin+vindesine+mitomycin C regimen (RR: 9.36, 95% CI: 1.29–68.0, P=0.027) are associated with treatment-related death from chemotherapy. Pulmonary fibrosis identified on a plain chest X-ray film (RR: 165.7, 95% CI: 8.79–3122, P<0.001), the combination of cisplatin+irinotecan (RR: 120.5, 95% CI: 2.90–4993, P=0.012), advanced age (RR: 1.17, 95% CI: 1.002–1.37, P=0.047), and elevated lactate dehydrogenase (LDH) (RR: 10.4, 95% CI: 1.20–90.2, P=0.033) were also associated with treatment-related death from thoracic radiotherapy. The administration of mitomycin C in addition to cisplatin-based regimens for patients with lung cancer should be carefully considered.

Introduction

Lung cancer is one of the most common carcinomas, not only in Japan, but also in the USA and Europe 1, 2. In 1990, the estimated number of patients with lung cancer worldwide was approximately 850 000 cases [1]. Approximately 45 000 patients die from lung cancer each year and the mortality is still increasing in Japan [3]. Surgery is the most curative treatment for early stage non-small cell lung cancer (NSCLC), however, only 30% of patients with NSCLC receive curative resection [4]. Cisplatin-based chemotherapy offers survival benefit and symptom relief for patients with inoperable NSCLC [5]. Combination chemotherapy is established as a standard therapy for extensive stage small cell lung cancer (SCLC) and the combination of chemotherapy with thoracic radiotherapy has also become a standard therapy for limited stage SCLC [6]. Thus, throughout the world a huge number of patients with lung cancer receive chemotherapy and/or thoracic radiotherapy.

Pneumonia and sepsis during neutropenia are common complications from chemotherapy and are sometimes lethal. Only a few reports have focused on the risk of treatment-related death in cases of inoperable lung cancer; those studies analysed patients with SCLC who were treated as a part of a clinical trial 7, 8, 9. Radford and colleagues reported that 50 of 382 patients (13%) with SCLC developed severe or life-threatening sepsis and 20 patients (5%) died due to sepsis [8]. Stephens and colleagues reported that 10% of patients with SCLC died within 3 weeks of the start of chemotherapy, and half of these may have been treatment related [9]. In these reports, poor performance status (PS), age >50 years, three-drug regimen, white blood cell count ⩾10×10⁹/l, hepatomegaly, higher alkaline phosphatase (Alp) and elevated blood urea (BUN) were listed as risk factors of treatment-related death in the treatment of SCLC 7, 8, 9. Radiation pneumonitis is also a common toxicity after thoracic radiotherapy, and occasionally, it develops outside of the radiation field. Severe pneumonitis causes hypoxia and death. Several reports have proposed risk factors for radiation pneumonitis, but none have discussed the risk factors of treatment-related death from thoracic radiotherapy 10, 11, 12, 13.

The incidence and risk factors of treatment-related death in the treatment of lung cancer using chemotherapy and/or thoracic radiotherapy are not well understood. Here, we analysed the incidence and risk factors of treatment-related death in the treatment of both SCLC and NSCLC including not only clinical trials, but also clinical practice-based treatments. In our hospital, approximately 50% of patients with lung cancer received chemotherapy and/or thoracic radiotherapy. It has been estimated that approximately 850 000 new cases are diagnosed each year worldwide [2]. If the same proportion of lung cancer patients worldwide, as in our hospital, receive chemotherapy and/or thoracic radiotherapy every year, it could be estimated that approximately 400 000 lung cancer patients receive chemotherapy and/or thoracic radiotherapy every year. Thus, if the incidence of treatment-related death could be reduced, even if it is only by 1% per year, 4000 patient deaths due to the toxicity of their treatment might be avoided. Thus, we believe that this study could provide useful information for lung cancer patients and for physicians who treat such patients.