Real-world treatment patterns for patients 80 years and older with early lung cancer: a nationwide claims study

Lung cancer is the leading cause of cancer-related deaths in South Korea, although the 5-year survival rate has increased from 11.3 to 25.1% [1]. Lung cancer develops two or three times more frequently in individuals 70 years and older than in younger individuals, and it is the leading cause of cancer-related deaths in those aged 80 years and older [2]. Because the incidence of cancer is expected to increase among the elderly [3], it will be important to improve patient survival, especially in this population.

Surgical resection for early-stage lung cancer is the best treatment to improve survival. However, elderly patients (> 80 years old) with comorbidities do not easily receive surgery due to concerns about surgery-related mortality and morbidity [4‐6]. Additionally, patients at this age often abandon the opportunity to cure early-stage non-small-cell lung cancer (NSCLC) due to the number of comorbidities, which increase proportionally with age [7, 8]. Stereotactic body radiation therapy (SBRT) is an alternative treatment modality that destroys lesions by delivering a high dose of radiation to the target site. SBRT showed a high local control rate comparable to lobectomy in early-stage NSCLC patients who were not eligible for surgery, with tolerable toxicities in patients with underlying emphysema and vascular diseases [9, 10]. The excellent local control and lower complication rates of SBRT may lead to changes in practical treatment patterns in patients with early-stage lung cancer who are not suitable for surgery or refuse treatment due to concerns over surgical complications. Toxicity issues associated with the treatment of early-stage NSCLC in the elderly remain, although previous studies have suggested that SBRT is safe and tolerable in elderly populations [11]. A recent study reported that the practice pattern and treatment outcomes have changed in elderly stage I lung cancer patients since the introduction of SBRT in the United States [12]. However, there was no report of annual survival outcomes for surgery and SBRT in a limited cohort of 80 years and older patients with early-stage lung cancer using a population-based database from South Korea.

This study evaluated the trends in practice patterns and survival outcomes over 9 years in patients 80 years and older diagnosed with early-stage lung cancer and compared the overall survival (OS) rates of those who underwent surgery and SBRT. We used a national claims database provided by the Health Insurance Review and Assessment service (HIRA).

In this population-based study, we uncovered temporal practice changes in patients 80 years and older diagnosed with early-stage lung cancer between 2008 and 2016 in real practice in South Korea. Although most patients received surgical resection (74.9%), there were significant changes in treatment trends at the time of the study period. Compared to 2008, the rate of surgical resection decreased gradually from 90.6 to 71.4%. However, the rate of SBRT increased over time from 9.4% in 2008 to 28.6% in 2016. The standard treatment for early-stage lung cancer is surgical resection with lobectomy [13]. However, in cases of poor surgical candidates with old age and comorbidities, it is not easy to perform surgical resection because of the high risk of complications. Since the introduction of SBRT for non-operable lung cancer due to comorbidities such as severe emphysema or heart disease [14], it has been adopted in clinical practice as an alternative to surgery accompanied by expected high risks of complications in patients with early-stage lung cancer. Practice pattern changes in elderly stage I NSCLC patients in which SBRT replaced surgery based on older age is now recognized in the United States [12]. Our data also suggest that SBRT increased its proportion substantially in elderly early-stage lung cancer patients in South Korea.

In our study, the median OS and 5-year survival rate in patients 80 years and older was 49.8 months and 44.8% with local treatment for early-stage lung cancer, respectively. A large population-based study analyzed 101,844 cases of lung cancer registered in the California Cancer Center for 4 years; stage 1 lung cancer was diagnosed in 19,702 patients. Of these, 7.3% of patients did not receive surgery or any type of radiation or chemotherapy. In this untreated group, the median OS and 5-year survival rate was 9 months and 6%, respectively [15]. In particular, lung cancer in the elderly has shown increased mortality, regardless of stage [16]. Therefore, screening for lung cancer with low-dose computed tomography [17] may be considered for those older than 80 years to reduce mortality in at-risk patients, although the evidence is not sufficient. However, there are concerns that old age makes treatment decisions difficult due to increased comorbidities and decreased physiologic performance. In advanced-stage NSCLC, a substantial number of patients were not treated and older patients were more likely to not receive any treatment [18]. Patients 80 years and older are less likely to be subjected to local treatment compared to younger patients; this results in a poor survival rate in older patients with lung cancer. Although we were unable to compare the survival rate between patients who received treatment and those who did not, definitive local treatment with surgery or SBRT should be performed in patients 80 years and older to increase the survival rate.

Our study shows that surgical resection had higher survival outcomes compared with SBRT after adjusting for variables using these population-based data. However, this result does not represent the head-to-head comparison of OS between the two treatment modalities. SBRT is generally adopted for patients who are not surgically suitable. In any nationwide claims study, selection bias is inevitable because it is impossible to obtain information regarding stage I or II cases, Eastern Cooperative Oncology Group (ECOG) performance, and factors related to surgical risk. Comorbidities increase with age and are the major reason for elderly patients with increased comorbidities to be precluded from surgical resection in cases of early-stage lung cancer [19]. Published surgical mortality and morbidity rates in patients 80 years and older range from 2.0 to 3.6% and 8.4 to 46%, respectively [4, 5, 20]. Although the Radiation Therapy Oncology Group confirmed SBRT as an alternative treatment modality for inoperable early-stage lung cancer [9], there are concerns about the safety and efficacy of SBRT in patients 80 years and older. A multicenter study reported that grade 2 radiation pneumonitis (RP) developed in 34.5% of patients and non-RP related toxicity was reported in 12.1%; however, no grade 4 or 5 toxicity was reported in the group who underwent SBRT for early-stage lung cancer (median age, 84.9 years) [21]. Another study regarding the safety of definitive SBRT in patients more than 80 years old suggested its high efficacy and tolerability [11]. Of interest, a prior study that investigated the effectiveness of SBRT at a single center reported that the OS rates for patients over 75 years were 86% at 1 year, 57.5% at 3 years, and 39.5% at 5 years [22], which is comparable to the survival rates of 79.1% at 1 year, 61.3% at 3 years, and 32.4% at 5 years in our patients who underwent SBRT.

Local relapse or distant metastasis is an important issue in early-stage lung cancer with definitive local treatment. The overall recurrence (local/regional/distant) reported in previous studies for elderly patients with SBRT was 30–40% [21, 23]. Therefore, the 8–9% rate of systemic chemotherapy after 3 months of local treatment in our study may be relatively low, in which surgery or SBRT revealed no significant difference in systemic chemotherapy after 3 months of definitive local treatment. This relatively low chemotherapy rate may be explained by old age-related comorbidities or poor ECOG performance. Although it should not be interpreted as a parameter of recurrence, we may indirectly estimate the rate of systemic chemotherapy as treatment failure after local control and suggest similar rates of treatment failure between the two modalities.

Our study has several limitations. A major limitation of nationwide claims data is the lack of clinical information. We could not obtain information related with performance status, underlying lung function with chronic obstructive pulmonary disease, interstitial lung disease, and accompanied comorbidities. Second, lung cancer stage (I or II) was not clearly divided in this study. The HIRA database contains only information related to claims covered by national insurance. Therefore, it was impossible to stratify the lung cancer staging according to the clinical information. Based on the claims database, we could retrieve early-stage lung cancer data only indirectly using patients who underwent surgical resection only or stereotactic body radiation therapy (SBRT) with no neoadjuvant treatment or adjuvant chemotherapy within 3 months after local treatment. Finally, local recurrence and distant metastasis could not be measured. Only data regarding chemotherapy 3 months later could be obtained; therefore, the rate of treatment failure may be underestimated.

The rate of SBRT increased gradually in patients 80 years and older with early-stage lung cancer, in contrast with the decrease in surgical resection. Surgical resection or SBRT is a reasonable treatment choice for elderly patients with early-stage lung cancer and should be adopted in light of multifactorial decision making.