Differences in Outcome and Toxicity Between Asian and Caucasian Patients With Lung Cancer Treated With Systemic Therapy

Ross A Soo; Tomoya Kawaguchi; Marie Loh; Sai-Hong I Ou; Marie P Shieh; Byoung-Chul Cho; Tony S Mok; Richie Soong

Disclosures

Future Oncol. 2012;8(4):451-462. 

In This Article

Differences in Efficacy

Small-cell Lung Cancer

In a retrospective population-based study of patients with SCLC in three southern Californian counties, Ou and colleagues reported the overall survival of Asian and caucasian patients with extensive-stage SCLC to be 7 and 5 months, respectively (p = 0.0438).[17] On multivariate analysis, Asian ethnicity was an independent favorable prognostic factor for overall survival (hazard ratio [HR]: 0.785; 95% CI: 0.657–0.938; p = 0.0076) (Table 1).[17] In a second study, an analysis of Japan Clinical Oncology Group (JCOG) 9511 and Southwest Oncology Group (SWOG) 0124 was performed.[18] These two studies were Phase III trials of cisplatin plus etoposide versus cisplatin plus irinotecan for SCLC. Eligibility criteria and treatment regimens were similar between the two studies. A pooled comparative outcome analysis of individual patient data reported that Japanese patients had higher overall survival compared with US patients (12.8 vs 9.9 months, respectively; p < 0.001), and higher overall response rate (87 vs 60%, respectively; p < 0.001).[18] In Japanese and US patients treated with the reference treatment arm, cisplatin and etoposide, overall survival was similar (9.4 vs 9.1 months respectively), whereas the response rate was higher in the Japanese patients (68 vs 57% respectively; p = 0.01)

Non-small-cell Lung Cancer

Several cancer registry studies have reported the association between survival and Asian ethnicity in NSCLC (Table 1). In a study of the California Cancer Registry from 1989 to 2003, 19,702 patients with stage I NSCLC were analysed. Asian ethnicity was an independent factor that decreased the risk of death in stage IA (HR: 0.813; 95% CI: 0.700–0.944; p = 0.0066) and IB NSCLC (HR: 0.803; 95% CI: 0.718–0.899; p = 0.0001).[19]

In a population-based study of NSCLC cases from the cancer surveillance programs of a Regional California Cancer Registry covering three southern Californian counties, 20,140 NSCLC patients with known smoking status were analyzed. Asians had the longest overall survival among the four major ethnicities (p = 0.0001) on univariate analyses. When analyzed according to AJCC stage, Asians had longer survival than non-Asians in stage III (12 vs 9 months respectively; p = 0.0136) and IV disease (6 vs 4 months respectively; p < 0.001) (Table 2). In multivariate analyses, Asian ethnicity was a favorable prognostic factor for overall survival (vs non-Asian; HR: 0.861; 95% CI: 0.808–0.918; p = 0.0001), independent of factors such as age, gender, smoking status, AJCC stage, histology, histologic differentiation, period of diagnosis and treatment.[5] Importantly, in this study, the prognostic impact of Asian ethnicity was shown to be independent of smoking status, which is a well-recognized prognostic factor.[20]

In a comparison between a hospital-based cancer registry from Korea of 4622 patients and a Regional California Cancer Registry of 8846 patients, Korean patients had a longer survival than US caucasian patients (HR: 0.869; 95% CI: 0.826–0.913; p < 0.0001).[21] Furthermore, this finding was independent of smoking status, confirming the results of the previous study by Ou et al..[5] The HR for overall survival in Korean patients compared with caucasian patients on univariate analysis for stage I disease was 0.618 (95% CI: 0.543–0.705; p < 0.001), stage II was 0.836 (95% CI: 0.867–1.016; p = 0.0723), stage III was 0.772 (95% CI: 0.712–0.836; p < 0.001) and stage IV was 0.846 (95% CI: 0.800–0.894; p < 0.001).

In a study from The National Hospital Study Group for Lung Cancer in Japan where 15,185 Japanese patients were compared with data from 13,332 patients from the Regional California Cancer Registry, patients of Japanese ethnicity had a longer survival than caucasian patients (HR: 0.937; 95% CI: 0.898–0.978; p = 0.0028).[22] When analysed according to AJCC stage, Japanese patients had a longer survival than caucasian patients with stage III (HR: 0.830; 95% CI: 0.789–0.873; p < 0.001) and with stage IV (HR: 0.955; 95% CI: 0.915–0.997; p = 0.0369).

In a study using the SEER Program database, 47,532 patients diagnosed with advanced-stage NSCLC between 1990 and 2003 were analyzed for survival differences.[23] Asian patients, in particular Chinese and Filipino, had a longer survival compared with caucasian patients (Table 1). The 1-year lung cancer-specific survival (95% CI) for caucasian, Chinese, Japanese and Filipino patients was 14.7% (14.3–15.0), 19.1% (16.4–22.0), 15.3% (12.5–18.4) and 18.3% (15.2–21.7), respectively.

Differences in sensitivity of Asian and caucasian patients to the EGF receptor (EGFR) tyrosine kinase inhibitors (TKIs) gefitinib and erlotinib have been well described. In a randomized Phase II study conducted in Europe, Australia, South Africa and Japan of two doses of gefitinib (250 vs 500 mg/day) in patients with advanced-stage pretreated NSCLC (the IDEAL 1 trial), the overall response rate was 18.4 and 19.0%, respectively. However, the response rate was higher for Japanese patients than non-Japanese patients (27.5 vs 10.4%, respectively; odds ratio [OR] = 3.27; p = 0.0023).[24] In a multivariate analysis, adenocarcinoma was predictive of response, with the odds of responding to gefitinib approximately 3.5-times higher for patients with adenocarcinoma than for patients with other tumor histologies. This reflects previous reports suggesting that EGFR mutations are more common in adenocarcinomas versus cancer of other histologies.[25]

A retrospective analysis of the IDEAL 1 study showed that in Japanese patients, gefitinib doses of 250 and 500 mg/day were associated with median survival rates of 13.8 and 11.2 months, respectively,[26] compared with 7.6 and 8.0 months, respectively, in the overall population.[24] In a randomized Phase III study of erlotinib versus best supportive care (the BR21 study), subset analysis revealed a higher response rate (18.9% vs 7.5%, respectively; p = 0.02) and overall survival (HR: 0.7; 95% CI: 0.5–0.9; p = 0.01) in patients of Asian origin compared with non-Asians.[27] In the ISEL trial, patients with advanced stage NSCLC were randomized to gefitinib 250mg daily or placebo; no significant difference in survival was seen between gefitinib and placebo at 5.6 months and 5.1 months, respectively (HR: 0.89; 95% CI: 0.77–1.02). However, in a preplanned subset analysis, an improvement in overall survival was seen with gefitinib compared with placebo in patients of Asian origin (9.5 vs 5.5 months, respectively; HR: 0.66; 95% CI: 0.48–0.91; p = 0.01).[28]

The higher frequency of activating sensitizing EGFR mutations in Asian compared with non-Asian patients (30 vs 8%) is thought to explain the increased response to EGFR TKIs in this ethnic group.[13] Alternatively, KRAS mutations, which occur in approximately 20–30% NSCLC patients,[29,30] have been reported to be associated with poor response to EGFR-tyosine kinase inhibitors. KRAS mutation rates tend to be higher in current or former smokers than never smokers. Mutations are also more common among adenocarcinomas than other histologies, with no significant difference between males and females.

KRAS mutations are less common in Asians than in caucasian patients. In the largest report series from Japan by Noda et al., the KRAS mutation rate was only 8% (33 out of 410 patients) in NSCLC,[31] whereas Wu et al. reported a KRAS mutation rate of 3.80% in 237 non-small lung cancers and 5.03% in adenocarcinomas among Taiwanese patients.[32]

In a meta-analysis examining the association between KRAS mutations and tumor response in NSCLC patients treated with EGFR-TKIs, the objective response rate of patients with mutant KRAS was 3% (6/210), whereas the overall response rate of NSCLC patients with wild-type KRAS was 26% (287/1125). The relative risk ratio (RR) was 0.29 (95% CI: 0.18–0.47; p < 0.01).[33] In the subgroup analysis, by Asian-derived population or white-derived population, the pooled RR was 0.22 (95% CI: 0.07–0.63; p = 0.01) and 0.31 (95% CI: 0.17–0.54; p < 0.01), respectively. These findings were echoed in another systematic review and meta-analysis assessing KRAS mutations as a candidate predictive biomarker for anti-EGFR-targeted therapeutic strategies in metastatic colorectal cancer and NSCLC.[34] The authors stratified the NSCLC studies by ethnicity (Asian vs white) and found Asian patients to be less sensitive to KRAS mutations than white patients, although these findings were marginally statistically nonsignificant.[34]

The efficacy and toxicity of crizotinib, an ALK small-molecule inhibitor[35] has been reported recently in an expanded cohort of a Phase I trial of patients harboring the oncogenic fusion genes EML4 and ALK.[36] The efficacy of crizotinib in ALK-positive NSCLC patients according to Asian and non-Asian ethnicity in this study has been recently reported in Abstract form.[37] Most of the Asian patients from this Phase I trial were Korean or Japanese. A higher overall response rate to crizotinib was seen in Asian (n = 29) than non-Asian patients (n = 53) at 76 versus 47%, respectively. These preliminary results should be interpreted with caution given the small sample size and retrospective nature.

Multiple studies of patients with advanced-stage NSCLC treated with first-line chemotherapy have reported that Asians had a numerically longer survival compared with caucasian patients. In a comparison across studies using the same regimen, studies have suggested a numerically longer survival in Japanese studies compared with caucasian studies. In studies of cisplatin and gemcitabine, the overall survival was 14.8 months for Japanese patients and 8.6–10.9 months for trials conducted in Europe and the USA. The overall survival for Japanese versus US/European studies for carboplatin and paclitaxel was 12.3 months and 7.8–11.0 months, respectively, and for cisplatin/vinorelbine was 11.4 months and 8.6–10.1 months, respectively.[38]

Further evidence for survival differences can be ascertained from Japanese and US NSCLC chemotherapy studies using a common study protocol. In this approach, patients from Japan and the USA were treated in three separate prospective Phase III studies in advanced-stage NSCLC (JMTO LC0003, FACS and S0003) with its own comparator regimen, but were linked by a common treatment arm of carboplatin and paclitaxel. In addition, study design, eligibility criteria and staging were similar. The median overall survival was 12 and 14 months for Japanese patients and 9 months for US patients (p = 0.0006).[39] Response rates were similar at 32 and 37% in the Japanese patients and 33% in the US patients (p = 0.55).

The AVAiL study was a randomized Phase III trial that assessed the efficacy and safety of chemotherapy with or without bevacizumab 7.5 and 15 mg/kg every 3 weeks. The progression-free survival was prolonged in both groups using bevacizumab 7.5 (HR: 0.75; 95% CI: 0.62–0.91; p = 0.003) and 15 mg/kg (HR: 0.82; 95% CI: 0.68–0.98), whereas no overall survival benefit was observed with either doses of bevacizumab.[40] In a preplanned exploratory subset analysis of the efficacy of bevacizumab in Asian patients, the progression-free survival in the 7.5 and 15 mg/kg groups was not statistically different compared with placebo, whereas overall survival was prolonged in the 7.5 mg/kg group (HR: 0.46; 95% CI: 0.22–0.97).[41]

In a global Phase III study of patients with advanced stage NSCLC positive for EGFR on immunohistochemistry (the FLEX study), overall survival was improved with the addition of cetuximab to cisplatin and vinorelbine when compared with the same chemotherapy alone with a median survival of 11.3 versus 10.1 months, respectively (HR: 0.871; 95% CI: 0.762–0.996; p = 0.044).[42] Subset analysis revealed Asian patients had a longer survival regardless of the treatment received, with a median survival of 19.5 (16.4–23.3 months) and 9.6 months (9.0–10.4 months) in caucasian patients. The frequency of favorable prognostic factors (female gender, never smoker, adenocarcinoma and ECOG performance status 0–1) was higher in Asian than caucasian patients. Furthermore, EGFR TKIs were used more frequently in Asian than caucasian patients (61 vs 17%, respectively).

More recently, the overall survival and response rates of Asian and caucasian patients from a meta-analysis of summary data from 191 randomized controlled trials published between 1975 and 2010 involving 48,369 patients (6806 Asian and 41,563 caucasian) identified through MEDLINE was reported. The median overall survival and overall response rate in Asian and caucasian studies for all chemotherapy regimens was 10.1 and 8.0 months (p < 0.001), and 32.2 and 25.9% (p < 0.001), respectively. The median overall survival in Asian and caucasian studies for monotherapy and platinum doublets was 9.9 and 6.8 months (p < 0.001) and 10.4 and 8.6 months (p < 0.001), respectively.[43] The difference in overall survival between Asian and caucasian patients was seen in the era before and after the introduction of EGFR TKIs. In studies published prior to the introduction of EGFR TKIs, the median overall survival in Asian and caucasian studies for all chemotherapy regimens was 9.1 versus 7.3 months (p < 0.001), respectively, and the overall response rate was 29.0 and 23.0% (p < 0.006), respectively. Overall survival was also longer in Asian compared with caucasian studies in the era following the introduction of EGFR TKIs, with a median survival of 11.0 and 8.9 months, respectively (Table 3). A greater proportion of known prognostic factors such as adenocarcinoma, female gender, performance status, smoking status and stage III were seen in Asian studies. Such imbalance in these recognized prognostic factors may contribute to the improvement in overall survival in Asian studies. In view of this, multiple linear regression analysis was performed and confirmed ethnicity was an independent and major factor for overall survival.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....