br Shorter OS in patients with higher number of
3.2. Shorter OS in patients with higher number of metastatic organs
The KaplaneMeier survival curves showed a significant as-sociation between OS and the M factor status (p¼0.0306) (Fig. 1A). The number of metastatic organs was also signifi-cantly related to OS (p¼0.0001) (Fig. 1B); the median OS was 393 days in patients with two or fewer metastatic organs and
Please cite this article as: Kanaji N et al., Association of specific metastatic organs with the prognosis and chemotherapeutic response in patients with advanced lung cancer, Respiratory Investigation, https://doi.org/10.1016/j.resinv.2019.06.004
Table 1 e Patient and tumor characteristics.
Smoker/never smoker/unknown
300/98/2
Pack-years, average (range)
44 (0e225)
ECOG Performance Status
Adenocarcinoma
228
Squamous cell carcinoma
61
Non-small cell carcinoma, NOS
28
Others
14
Small cell carcinoma
69
Driver gene mutations
Median (days)
345
First-line therapy
Received
345
Not received; best supportive care
55
Progression-free survival
Median, days
147
Responses to first-line therapy
ALK: anaplastic lymphoma kinase, CR: complete response, ECOG:
Eastern Cooperative Oncology Group, EGFR: epidermal growth
factor receptor, NE: not evaluable, NOS: not otherwise specified, PD:
progressive disease, PR: partial response, RET: rearranged during
transfection, ROS1: c-ros oncogene 1, SD: stable disease.
206 days in patients with three or more metastatic organs. Both of the following cut-off points were also significant: one versus two or more organs (p¼0.0017) and three or fewer versus four or more organs (p¼0.0022).
The univariate analysis using the log-rank test revealed that older age, male sex, smoking history, poor performance status, the presence of ILD, SCLC histology, the absence of driver gene mutations, M1c but not M1a/b, and a high number of metastatic organs ( 3) were associated with shorter OS (Table 2).
3.3. Difference in survival according to the sites of metastatic organs
The most frequent metastatic sites in this series of patients with lung cancer were within the chest cavity, as observed in 202 patients (51%); this was followed by bone (n¼179, 45%) and Nifurtimox (n¼108, 27%) metastases (Table 2). However, none of these sites were associated with OS (Table 2). Metastases in the liver (n¼76, 19%), adrenal gland (n¼73, 18%), muscle (n¼19, 5%), and skin (n¼12, 3%) were associated with shorter OS. When patients with NSCLC and SCLC were assessed sepa-rately, liver, adrenal gland, and muscle metastases were associated with a poor prognosis in patients with NSCLC, whereas bone and liver metastases were associated with a poor prognosis in patients with SCLC (Supplementary Table S1). The multivariate analysis using a Cox proportional-hazards model showed that older age, poor performance status, the presence of ILD, the absence of driver gene muta-tions, liver metastasis, and muscle metastasis were indepen-dently associated with shorter OS (Table 2, Fig. 2). The median OS was 207 and 120 days in patients with liver and muscle metastases, respectively. Other metastatic sites were the gastrointestinal tract (n¼10), peritoneum (n¼9), pancreas (n¼9), cerebrospinal fluid (n¼4), bone marrow (n¼4), and others. When limited to an M1b status, bone metastasis was associated with better OS and skin metastasis was associated with poor OS (Supplementary Table S2).
Because a higher number of metastatic organs was asso-ciated with shorter OS in the univariate analysis but not in the multivariate analysis, the difference in OS among the meta-static organ sites might simply depend on the number of metastatic organs. We observed a high number of metastatic organs in all patients who had metastatic organs (Fig. 3A). In particular, in patients with muscle or skin metastasis, the total number of metastatic organs was more than four (4.3 or
Fig. 1 e KaplaneMeier curves of overall survival (OS) in patients with stage IV lung cancer dependent on the status of the (A) M factor and (B) number of metastatic organs at the time of diagnosis of lung cancer.
Please cite this article as: Kanaji N et al., Association of specific metastatic organs with the prognosis and chemotherapeutic response in patients with advanced lung cancer, Respiratory Investigation, https://doi.org/10.1016/j.resinv.2019.06.004
Table 2 e Risk factors associated with OS.
Characteristic
n
Median OS, days
Univariate analysis
Multivariate analysis
p-value
Age
Sex
Smoking history
Never smoker
98
ECOG Performance Status