Favorable Trends of Lung Cancer Mortality-to-Incidence Ratios in Countries With High Computed Tomography Density

Background: Prognoses for lung cancer deteriorate dramatically with the progression of cancer stages. Therefore, early screening by techniques such as low-dose computed tomography (LDCT) is critical. However, the epidemiology regarding the association between the popularization of CT and the prognosis for lung cancer is not known. Methods: Data were obtained from GLOBOCAN and the health data and statistics of World Health Organization. MIRs and the changes in MIR over time ( 𝛿 MIR), which were calculated as the difference between MIRs in 2018 and 2012, were used to evaluate the correlation to CT density disparities via Spearman's rank correlation coecient. Results: Countries with zero CT density presented a relatively low incidence crude rate and a relatively high MIR in 2018 and a negative 𝛿 MIR. Conversely, countries with CT density over 30 had a positive 𝛿 MIR. The CT density was signicantly associated with human development index (HDI) score and MIR in 2018 but demonstrated no association with MIR in 2012. The linear correlation between CT density and 𝛿 MIR also shows a signicant association. Conclusion: CT density was signicantly associated with MIR in 2018 and with 𝛿 MIR, indicating favorable clinical outcomes in countries with popularization of CT.


Introduction
Lung cancer used to be a rare disease; however, since the start of the 21st century, it has become the cancer with the highest incidence and mortality worldwide. In 2018, the number of new cases of lung cancer was as high as 2,093,876 and deaths due to lung cancer totaled 1,761,007 [1]. These data are strikingly high and have aroused public concern. Moreover, mortality rates in 2018 closely paralleled the incidence rate in lung cancer worldwide, meaning a poor treatment outcome for lung cancer after diagnosis [1]. Lung cancer is usually diagnosed in advanced stages [2], and the prognosis at this period is extremely poor. Therefore, early diagnosis is a true public concern.
If lung cancer can be diagnosed early, the ve-year survival rate can rise to 60-80%, much higher than in stage 3 (16%) and stage 4 (less than 10%) [3]. Unfortunately, early diagnosis becomes quite di cult because lung cancer in its early stage is asymptomatic and its detection depends on image surveys, such as X-ray or CT. However, compared to CT screening, the use of X-rays to nd abnormalities in lung cancer, and especially small lesions, if far more di cult. In 90% of early lung cancer cases, a misdiagnosis occurs when X-rays are used [4], due to the di culty in distinguishing lung lesions from bones, pulmonary vessels, mediastinal structures, and other complex anatomical structures [4]. Conversely, CT has high sensitivity and can distinguish 83% to 91% of these lesions [5].
Low-dose computed tomography (LDCT), as the name suggests, allows screening at a lower radiation dose (1.0-1.4mSV) compared to traditional CT [6]. Moreover, related research showed no signi cant differences between LDCT and CT when used for lung cancer screening, as the concordance rate for the diagnoses was approximately 80% [7]. CT mainly detects lung nodules and establishes baseline screening for lung cancer [8]. However, the CT density, de ned as the total density of CT facilities per million population, varies dramatically from region to region. CT facilities require abundant funds, so their construction might be in uenced by the economic development of a given country. Based on the important role of LDCT in lung cancer prognosis and the regional differences in CT density, we considered that CT density might affect the worldwide mortality-to-incidence ratios (MIRs). Many previous studies have focused on the effectiveness of LDCT by analyzing the correlation between regular LDCT screening and lung cancer mortality rates, whereas few studies have explored the real situation regarding the availability of LDCT worldwide. Here, we have provided a comprehensive view of the relevance of CT availability to lung cancer prognosis according to regions by analyzing the association between CT density and MIRs.

Materials And Methods
Lung cancer (ICD-10 C33-34) epidemiological data were obtained from the GLOBOCAN database (https://gco.iarc.fr/today/) estimates from 2012 and 2018 for 185 countries. The human development index (HDI) was obtained from the United Nations Development Programme, Human Development Report O ce (http://hdr.undp.org/en). Data for CT density for 2013 were obtained from the Global Health Observatory data repository (https://www.who.int/data/gho). The data on health expenditures, including per capita current health expenditure (CHE) and current health expenditure on gross domestic product (CHE/GDP) (ratio of CHE to the % of GDP), were obtained from the World Health Statistics database (https://www.who.int/gho/publications/world_health_statistics/en/). MIR was de ned as the ratio of the crude rate of mortality to the crude rate of incidence, as previously described [9][10][11][12]. The The exclusion criteria for country selection included data missing for the CT density (N = 60), missing data for MIR/HDI/CHE (N = 3), and outliers for the CT density (N = 2). A total of 115 countries were included in the nal analysis.
The associations between the MIR, MIR, and other factors among the various countries were estimated using a Spearman's rank correlation coe cient calculated using SPSS statistical software version 15.0 (SPSS, Inc., Chicago, IL). Values of p < 0.05 were considered statistically signi cant. Scatterplots were generated using SigmaPlot. Table 1 demonstrates the scores and ranks for the human development index and the CT density according to countries. Countries with a CT density less than 0.01 are the African Republic, Guinea-Bissau, Guinea, and Vanuatu. As expected, these countries also rank low in terms of HDI (187, 174, 176, and 122, respectively). By contrast, countries with CT density higher than 20 all rank in the top 50, except for Lebanon. The highest CT density is in Iceland (39.45), and the lowest CT density are in Guinea-Bissau and Guinea (0.00).

Result
CT density and incidence crude rates, MIR, and MIR disparity of lung cancer Those countries with zero CT density, namely the African Republic, Guinea-Bissau, Guinea, and Vanuatu, also have a incidence crude rate lower than 10 (2.2, 2.8, 2.4 and 8.0, respectively). Furthermore, their MIR in 2018 was higher than 0.9 and the MIR values were all negative. In terms of MIR, countries with a CT density over 30, such as Iceland, Greece, and South Korea, have a positive MIR (0.15, 0.12, and 0.11, respectively). Conversely, Vanuatu had an extremely low MIR (-0.14) and the highest MIR in 2018 (1.06) and had a CT density of zero.
The association between CT density, MIR in 2012 and 2018, and MIR.

Discussion
The previous large National Lung Screening Trial (NLST) with an enrolment of 53,454 persons, focused on the effect of routine LDCT screening on mortality reduction, but only concluded that ever-smokers were at high risk [14,15]. Our study, also based on the NLST, generally explored the association between CT density distribution and the prognosis of lung cancer and revealed the concordance between the geographical distribution of actual CT density and prognosis of lung cancer. Adequate CT facilities might better meet the needs of people eligible for routine screening, thereby improving the prognosis of lung cancer.
According to the NLST, the LDCT can reduce lung cancer mortality by 20% compared to X-rays, and follow-up European studies also support the effectiveness of LDCT [16,17]. LDCT mainly detects nodules in the lung. The size, growth rate, morphology, and location of the nodules are references used for malignant or benign judgments. Related research shows that most lung cancer is con rmed in large nodules, whereas lung nodule counts have yet to be proved to determine malignancy [18]. Compared to the high misdiagnosis rate and the increased cost and time-consuming shortcomings of MIRs, regular LDCT screening might be a more practicable form of radiography for close tracking of new large nodules [4,19]. Due to the effectiveness of LDCT screening, the United States recommends that current and former smokers aged 55 through 80 years with a 30-year history of smoking receive routine LDCT screening [20].
The recommendation for LDCT screening of high-risk people was based on NLST research and was announced in 2011 [14,20]. This might explain the lack of a signi cant association between CT density and MIR in 2012 in our analysis and the subsequent signi cant association in 2018. Furthermore, CT density is also signi cantly associated with HDI. According to previous research on the cost-effectiveness of CT, which has an incremental cost-effectiveness ratio (ICER) of £10,069 per quality-adjusted life year (QALY) [21]. This number might be an imposing burden for low-HDI countries. In other words, the CT density might also re ect the amount of money invested in medical care.
To the best of our knowledge, our study is the rst to aim at identifying an association between CT density and the MIRs of lung cancer. However, this study still has some limitations. One is that only countries with a CT density recorded by the WHO were included in our study. Another is that the level of CT density might not be representative of the people that actually receive routine screening. A third limitation is that previous research showed no signi cant association between LDCT and the decrease in lung cancer mortality rates in males, meaning that the study might ignore sexual differences [5]. Despite these limitations, our study still demonstrates that the prognosis of lung cancer might improve as the CT density rises.

Abbreviations
Low-dose computed tomography, LDCT Mortality-to-incidence ratios, MIRs Human development index, HDI Current health expenditure, CHE Current health expenditure on gross domestic product, CHE/GDP MIR over time, MIR National Lung Screening Trial, NLST Incremental cost-effectiveness ratio, ICER

Con icts of Interest
The authors declare no con ict of interest.  Tables   Table 1. Summary of human development index, CT density, cancer incidence, cancer mortality, and mortality-to-incidence ratio in lung cancer of selected countries.  Figure 1 The human development index of (A) all selected countries (n = 115), and of countries with (B) HDI score < 0.70 (n = 56), and (C) HDI score ≧ 0.70 (n = 59), are signi cantly associated with CT density.