TNM Stages Inversely Correlate With the Age at Diagnosis in ALK-Positive Lung Cancer

Wenfang Tang; Yuanyuan Lei; Jian Su; Chao Zhang; Rui Fu; Jin Kang; Honghong Yan; Xuening Yang; Haiyan Tu; Yilong Wu; Wenzhao Zhong

Disclosures

Transl Lung Cancer Res. 2019;8(2) 

In This Article

Results

Patient Characteristics

In the ALK-positive cohort, 411 (4.9%) eligible patients were identified, including 127 patients with stage I–IIIa disease and 284 patients with stage IIIb–IV disease (Figure 1). Overall, 383 (93.2%) patients had histologically confirmed adenocarcinoma. Of the 411 eligible patients, there was almost an equal proportion of females [n=204 (49.6%)] and males [n=207 (50.4%)] in our study, and 321 (78.1%) patients in our cohort had never smoked. The ECOG scores of the patients were primarily low (score =0–1) [n=380 (92.5%)]. Moreover, the majority of ALK-positive patients had an absence of brain metastasis at the initial diagnosis [n=349 (84.9%)]. Of the treatment strategies, 129 (31.4%) patients underwent surgery, 180 (43.8%) patients received targeted therapy, and 150 (36.5%) patients were treated with chemotherapy or/and radiotherapy (Table 1).

The median age at diagnosis of the patients included in ALK-positive cohort was 51 years (range, 24–82 years). As shown in Table 1, young patients with ALK-positive lung cancer were associated with a high likelihood of being female (P=0.048), having histological adenocarcinoma characteristics (P=0.004), and exhibiting low ECOG scores (P=0.006). In addition, young patients more frequently had diseases in the T3/4 stage (P=0.014), lymph node metastases (P=0.011) and distant metastasis (P=0.015) (Table 1) than old patients.

There were 122 (5.6%) patients in the KRAS-mutant cohort who had a median age at diagnosis of 59 years (range, 40–88 years), including 46 patients with stage I–IIIa disease and 76 patients with stage IIIb–IV disease (Figure 1). As shown in Table 1, there were no significant differences in the clinical characteristics between the various age groups.

Association Between age at Diagnosis, Frequency and TNM Stages

Among all patients with ALK-positive lung cancer, the mean age at diagnosis decreased steadily with more advanced clinical stages [I/II vs.III vs. IV, mean age ± standard deviation (SD): 54.7±11.4 vs. 52.0±10.5 vs. 49.7±11.6 years]. There was a significant difference in age at diagnosis between the various clinical stages (P=0.002), and the age at diagnosis inversely correlated with the clinical stages (P<0.001) (Figure 2A). Moreover, these associations also existed for T stages (T1 vs. T2 vs. T3/4, mean age ± SD: 53.2±10.7 vs. 52.7±11.0 vs.49.4±12.1 years), N stages (N0/1 vs. N2 vs. N3, mean age ± SD: 53.6±11.8 vs. 51.7±11.0 vs. 49.2±11.4 years), and M stages (M0 vs. M1, mean age ± SD: 53.4±11.0 vs. 49.7±11.6 years). Significant differences were also observed in age between various T, N, and M stages (T stages: P=0.003; N stages: P=0.004; M stages: P=0.001) and the age at diagnosis inversely correlated with the T, N, and M categories (T stages: P=0.001; N stages: P=0.001; M stages: P=0.001) (Figure 2B,C,D). However, the KRAS-mutant patients did not demonstrate similar characteristics to those of ALK-positive patients. The mean age at diagnosis of KRAS-mutant patients was 63.0±8.9, 58.1±9.6, 59.4±10.5 years for stage I/II, III, IV disease, respectively. There was no significant difference in age at diagnosis between various clinical stages (P=0.064) and there was no significant inverse correlation between age at diagnosis and clinical stages (P=0.084) (Figure 2A). Although the age at diagnosis showed a significant inverse correlation with the N categories (P=0.010), the correlation was not significant in the T (P=0.887) or M categories (P=0.152) (Figure 2B,C,D).

Figure 2.

The correlations between age at diagnosis, frequency, and TNM stages for ALK-positive and KRAS-mutant lung cancer. (A,B,C,D) Mean ages at different TNM stages; P*, as determined by Wilcoxon rank sum tests, which demonstrated the differences in age between various TNM stages; P#, as determined by Spearman correlation tests, which demonstrated the correlation between age at diagnosis and the TNM stages; (E) percentage of ALK-positive and KRAS-mutant lung cancer being stage IIIb–IV disease in the relative age groups; P, as determined by linear regression tests, which demonstrated the trend of percentage of stage IIIb–IV disease between the various age groups; (F) frequency of ALK-positive and KRAS-mutant lung cancer being stage I–IIIa and IIIb–IV diseases in the total population. ALK, anaplastic lymphoma kinase; KRAS, Kirsten rat sarcoma viral oncogene homolog.

Furthermore, the proportion of ALK-positive lung cancer that was stage IIIb–IV disease decreased steadily as the age groups became older (84.8% vs. 73.6% vs. 66.1% vs. 56.9%) (F=338.4; P=0.003). However, the KRAS cohort did not show this significant linear relation (F=10.6; P=0.190) (Figure 2E). In this study, the total frequency of ALK rearrangements was approximately 4.9%. We found that the frequency of ALKrearrangements in patients with stage IIIb–IV disease was much higher than that of the patients at stage I–IIIa disease (6.1% vs. 3.4%, P<0.001). However, the frequency of the KRAS mutation in patients with stage I–IIIa and IIIB–IV disease were similar at 5.5% and 5.6%, respectively (P=0.924) (Figure 2F).

Survival Analysis

The median follow-up time was 13.7 months (range, 0.1–95.8 months), and the last follow-up was recorded on August 10, 2018. In the ALK-positive cohort, 112 (27.3%) patients died during follow-up. The 5-year OS rates were 40.5%, 47.8%, 44.8% in the <40 years, 40–49 years and 50–59 years groups, respectively. Among the ALK-positive patients aged 60 years or older, the 5-year OS rate reached 65.6%. There were statistically significant differences in OS between the ≥60 years group and <40 years group (P=0.048) and between the ≥60 years group and the 50–59 years group (P=0.041). Although the OS between ≥60 years group and the 40–49 years group was not significantly different (P=0.135), the patients in the ≥60 years group were associated with a trend toward better survival compared to the other age groups (Figure 3A and Table 2). However, in the KRAS cohort, there were no significant differences between the three age groups (40–49 vs. 50–59 years, P=0.052; 40–49 vs. ≥60 years, P=0.205; 50–59 vs. ≥60 years, P=0.451) (Figure 3B and Table 2).

Figure 3.

Survival analysis stratified by the different age groups. ALK, anaplastic lymphoma kinase; KRAS, Kirsten rat sarcoma viral oncogene homolog.

For the ALK-positive cohort, multivariable analyses revealed that non-intracranial metastatic disease [hazard ratio (HR): 2.87; P<0.001] and the presence of brain metastases at diagnosis (HR: 3.72, P<0.001) were associated with poor survival, as was the presence of high ECOG scores (HR: 3.97; P<0.001). Furthermore, non-intracranial metastatic disease (HR: 4.90; P<0.001) and the presence of brain metastases at diagnosis (HR: 4.76, P<0.001) were also prognostic factors in KRAS-mutant patients (Table 3).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....