The relationship between urologic cancer outcomes and national Human Development Index: trend from 2012 to 2018

Objectives To describe the influence of the socioeconomic development on worldwide age-standardized incidence and mortality rates, as well as mortality-to-incidence ratio (MIR) and 5-year net survival of urologic cancer patients from 2012 to 2018. Methods The HDI values were obtained from the United Nations Development Programme, data on age-standardized incidence/mortality rates of prostate, bladder and kidney cancer were retrieved from the GLOBOCAN database, 5-year net survival was provided by the CONCORD-3 program. We then evaluated the association between incidence/MIR/survival and HDI, with a focus on geographic variability as well as temporal patterns during the last 6 years. Results Urologic cancer incidence rates were positively correlated with HDIs, and MIRs were negatively correlated with HDIs. Prostate cancer survival also correlated positively with HDIs, solidly confirming the interrelation among cancer indicators and socioeconomic factors. Most countries experienced incidence decline over the most recent 6 years, and a substantial reduction in MIR was observed. Survival rates of prostate cancer have simultaneously improved. Conclusion Development has a prominent influence on urologic cancer outcomes. HDI values are significantly correlated with cancer incidence, MIR and survival rates. HDI values have risen along with increased incidence and improved outcomes of urologic caner in recent years.

positively with HDIs, solidly confirming the interrelation among cancer indicators and socioeconomic factors. Most countries experienced incidence decline over the most recent 6 years, and a substantial reduction in MIR was observed. Survival rates of prostate cancer have simultaneously improved. Conclusion Development has a prominent influence on urologic cancer outcomes. HDI values are significantly correlated with cancer incidence, MIR and survival rates. HDI values have risen along with increased incidence and improved outcomes of urologic caner in recent years.

Background
With the growing and aging population, cancer has been expected to rank as the leading cause of death and the most important barrier to increasing life expectancy across the world [1]. Urologists are the front line for the diagnosis and treatment of urologic malignancies, mainly including prostate, bladder, and kidney cancer [2]. According to Global Cancer Statistics 2018, bladder, prostate and kidney cancers have ranked the 3 rd , 3 10 th and 14 th most common tumors worldwide, respectively [1]. The global composition of urologic cancer patients has been continuously evolving due to multiple forces [3,2,1,[4][5][6][7][8][9].
Socioeconomic development is closely interconnected with public health [10]. Human Development Index (HDI) is the gold standard for the comparison of socioeconomic development, quantified by the composite measures of health, education, and economy [11]. Multiple studies had demonstrated that cancer outcomes were related with HDI [12,13]. Some identified there was a negative correlation between standardized mortality rates and HDI [14] but others verified no significant correlation [15]. This study aims to describe the influence of HDI values on global urologic cancer burden and outcomes, including incidence rates, mortality-to-incidence ratio (MIR) and 5-year survival rates, with a focus on the transition for the years 2012 through 2018.

Data sources
The incidence and mortality estimates of urologic cancer were originally extracted from the GLOBOCAN database (http://gco.iarc.fr) maintained by the International Agency for Research on Cancer (IARC). Data within 186 countries in 2018 and 175 countries in 2012 were incorporated. All indicators were presented in forms of age-standardized rates (ASRs) per 100,000 person-years. The ASRs were calculated according to the world standard population, allowing comparisons between populations without being influenced by differences in their age structures [1,16]. 4 HDI data for United Nations members during 2012-2018 period were available in the United Nations Development Programme (UNDP) database (http://hdr.undp.org/en/statistics). We further collected the 5-year net survival estimates of patients diagnosed with prostate cancer from the CONCORD-3 report which corresponded with patient status in year 2012 and 2018 [17]. Net survival is the cumulative probability of surviving up to a given time since diagnosis (e.g., 5 years) after correcting for other causes of death (background mortality) [17].

Statistical analysis
With the obtained incidence and mortality rates, we calculated the urologic cancer mortality-to-incidence ratio (MIR) (ie., cancer deaths divided by incident cancer cases).
Extreme values (0, 1 or > 1) were considered abnormal and were excluded from the analysis. To examine patterns in the MIR of urologic cancer by levels of socioeconomic development, we correlated the MIRs to the corresponding HDIs via linear and nonlinear regression. Linear regression fit was conducted to identify the existence of correlation.
Correlation was established with a significant p value in the nonparametric Spearman correlation test. Nonlinear regression was based on a modified "dose-to-response" model using the formula , HDI 50 refers to the half-maximal controlled HDI (equivalent to the HDI value at half-maximal MIR) and slope is a parameter indicating the steepness of the fitted curve. MIRs comparison among 4-tier HDI groups was analyzed via One-way ANOVA followed by Tukey-Kramer post hoc tests. We further examined the correlation of national incidence rates and 5-year net survival estimates with corresponding HDI, separately. In order to determine the effects of socioeconomic transitions on urologic cancer outcomes, we further compared the age-standardized MIR or 5-year net survival estimates in the year of 2012 and 2018 (paired t-test). A p value less than 0.05 was considered statistically significant. Statistical analysis and plotting were performed using Prism 7 (GraphPad, San Diego, CA).
Global geographical maps were depicted by TileMill (a GitHub software maintained by MapBox, Washington, DC), with map data sources from the Natural Earth database rendered by the Mapnik Library (https://mapnik.org/).

Overview of current global urologic cancer epidemiology
Development levels of countries were classified into 4 classes according to HDI values by the UNDP (Fig. 1a). The global age-standardized incidence and mortality rates of prostate, bladder and kidney cancer in 2018 were presented separately ( Fig. S1a-f).
Mortality-to-incidence ratios (MIRs) were calculated and their global distribution was depicted in the form of world maps ( Fig. 1b-d).
Prostate cancer ranked as the second most frequent cancer in men. The top countries with the highest incidence rates of prostate cancer were all in very-high HDI group (Fig. S1a), including Europe (e.g., Ireland, Estonia, Norway, Sweden, France, United Kingdom, etc.), North America (United States), Australia/ New Zealand, and Barbados. However, mortality rates did not follow those of incidence. The highest mortality rates fell mainly in countries with lower HDI (Fig. S1b), including the Caribbean and Africa. The lowest MIRs were achieved in highly developed countries (Fig. 1b), such as France, Ireland, Italy, Spain, United States. Whereas low-to-medium HDI countries owned the highest MIRs (Afghanistan, Guinea, Pakistan, Liberia). Similarly, bladder and kidney cancer was most common to happen in high-to-very high HDI regions (Fig. S1c, 1e), especially European countries. However, the top 10 countries with highest MIRs of bladder cancer were all from low-HDI group in Africa (Fig. 1c, 1d). The lowest MIR was from very-high HDI group.

The correlation between urologic cancer MIR and national HDI
The global MIR of prostate, bladder and kidney cancer in 2018 was 0.358, 0.251, and 6 0.410, respectively ( Fig. 1b-d). We found that as the level of national HDI increased, the corresponding urologic cancer MIR was relatively lower, with strong correlation (r < 0, p < 0.0001). We also applied nonlinear regression analysis on data, verifying the existence of a "dose-to-response" inhibitory effect between HDI values and MIRs ( Fig. 2a-

b, d-e, g-h).
The HDI values at half maximal MIR (HDI 50 ) of prostate, bladder and kidney cancer in 2018 was 0.639, 0.704 and 0.736, respectively.
We further compared the MIRs of urologic cancer among 4-tier HDI groups and clarified the persistent disparities associated with HDI levels (p < 0.0001, One-way ANOVA). Take prostate cancer in 2018 for example, the mean MIR in very-high HDI countries (0.224) was significantly lower than that in high-(0.424), medium-(0.522), or low-(0.641) HDI countries (p < 0.0001, Tukey's post hoc test; Fig. 2c). Similar results were obtained in other cancer sites (bladder, kidney) as well as data in 2012 (p < 0.0001, Tukey's post hoc test; Fig. 2f, i).

Association between incidence rates of urologic cancer and HDI
Since the fact that urologic cancers tended to happen more in high-to-very high-HDI countries, we also applied correlation analysis on the association between incidence rates and HDI. It was demonstrated that national incidence rates in urologic cancer all had strong correlation with corresponding HDIs via linear regression (r > 0, p < 0.0001; Fig. 3a-c).

The impact of HDI on 5-year survival of prostate cancer
The 5-year net survival rates were available in 57 countries for prostate cancer in Accordingly, the survival rates correlated inversely with national MIR (r = -0.730, p < 0.0001; Fig. 4c).  Fig. 2c). The survival rates generally increased, with only 7 out of 57 countries dropped more than 1% (p < 0.05) (Fig. 5b; Table 1).  Fig. 2f,   Fig. 5c). There was also only tiny fluctuation between MIR of 2012 and 2018 within each 8 specific HDI group (Fig. 2f).

Kidney cancer
The incidence of kidney cancer also reduced during 2012-2018 (337 860 vs 403 262). In the scatter diagram of HDI-MIR, current MIR of kidney cancer also shifted in the direction when comparing with 2012 (Fig. 5d). Remarkably, two regression lines were nearly parallel (r 2018 = -1.153, r 2012 = -1.231; p < 0.01). Meanwhile, MIRs of kidney cancer declined overall, across all development status (p < 0.01 for the low-, p < 0.0001 for the medium-, p < 0.001 for the high-, and p = 0.23 for the very-high-HDI group; Fig. 2i).

Discussion
Our study aims to clarify the latest epidemiology of urologic cancer and the contribution of national development to urologic cancer outcomes. Mortality-to-incidence ratio (MIR) and 5-year net survival both represent cancer outcomes. MIR is regarded as a quite useful surrogate indicator of oncology care effectiveness, which could be a comprehensive result of screening, diagnostic modality, treatment and follow-up [12,18]. Meanwhile, 5-year net survival might be labeled with more importance, since cancer patients who survive for a considerable time span can, in a way, be considered cured [17,19]. In the current study, we proved that all three urologic cancers MIR negatively, while incidence and survival rates positively correlated with HDI.

Prostate cancer
Incidence of prostate cancer was highest in countries with very-high HDI. The public recommendation and prevalence of early diagnostics for prostate cancer in more developed countries, by PSA testing and detection of latent cancer in transurethral prostatectomy or puncture biopsy, led to higher incidence rates. For example, the commercial availability of PSA testing from 1980s brought about the intensively use of the test and rapid growth in new cases, first in the United States and within a few years, in Europe, Australia/New Zealand, and Canada [1,20,21]. Another explanation could be attributed to age. Nearly 75% of new prostate cancer cases occurring in people aged over 85 years, and incidence of prostate cancer is directly correlated with age [22,15]. Since life expectancy is one of key elements of HDI, there is no doubt that countries with higher HDI had a greater prostate cancer incidence. Moreover, ethnic and genetic predisposition could also be blamed for prostate cancer morbidity. The rates are highest among men of African descent in the United States and the Caribbean [23]. That's why Barbados and Bahama topped in the incidence rates of prostate cancer with a relatively lower HDI within high-HDI group.
The correlation between cancer outcomes and HDI seemed to be driven by national inequalities in health care, resulting in deviations in treatment effectiveness. First of all, the widespread access to diagnostic services and screening tests in more developed countries leads to increased diagnosis at earlier stages of disease and better clinical outcomes. However, we could not deny the overdiagnosis associated with PSA screening.
Some new biomarkers sparing those who overdiagnosed are under development (e.g., PCA3 or TMPRSS-ERG fusions) [24]. Secondly, the delivery of urologic oncology care is susceptible to regional variation. Access to effective radiation equipment and neoadjuvant hormonal therapy is linked to a country's wealth. Advances in immunotherapy and robotic surgery, though promising, are not feasible or affordable for generalized application in settings with limited health care resources [25][26][27].
Multiple studies have reported that the detected cases of prostate cancer from 1990 to

Bladder Cancer
Higher incidence rates of bladder cancer could be observed in more developed countries, especially in Southern and Western Europe. Chemical or environmental exposures are major risk factors for bladder cancer, including smoking, obesity, alcohol drinking and redmeat consumption [31,32]. These risk factors have been reported by the World Health Organization as alarmingly high across Europe [33]. Among above, cigarette smoking is the most important environmental factor [31]. Smokers have a two-to five-fold increased risk of bladder cancer relative to never-smokers [34]. At present, smoking prevalence is highest among both genders in Europe [35]. More exposures to carcinogenic agents like aromatic amines in dye industry in Europe also leads to higher morbidity [36]. Other likely reason is widespread practice of initial assessment in more developed regions, consisting of voided urine cytology, cystoscopy, radiological investigation of the upper tracts for haematuria and other non-specific urinary symptoms [34,32]. MIR of bladder cancer appeared to be high in less developed regions. Higher-quality medical care and better health awareness were possessed in highly developed areas, as stated in Prostate Cancer.

11
The evolving burden of bladder cancers since 1990 could be attributed to following forces: population growth and aging, the improvement in diagnostic technology, and the changing risks of exposures like tobacco use and obesity [2]. While from 2012 to 2018, patients went in declines in both incidence and MIR of bladder cancer worldwide. Primary prevention of tobacco use is the most effective strategy for bladder cancer prophylaxis [22]. Currently, under the policies of tobacco control [37], smoking cessation is becoming common and may explain how the bladder cancer burden is becoming retrieving, especially in more highly educated and health-focused populations [32,22]. Meanwhile, decreasing incidence and earlier diagnosis also leads to stage migration to earlier stage disease. Another driver for cancer outcomes improvement could be due to improved endoscopic system for cystoscopic surveillance [38]; more transurethral bladder resection for non-invasive cancer [31]; radical cystoprostatectomy and anterior exenteration by robotic surgery with less invasive injury to patients [39]; better intravesical therapy, such as the Bacillus Calmette-Guerin (BCG) and updated chemotherapy [31]. Moreover, EGFR, ERBB2 and VEGF are validated targets for cancer therapy but still remain the subject of intense investigation so far [40,27].

Kidney cancer
The incidence of kidney cancer correlated positively with country-specific HDI. Risk factors such as obesity, hypertension and diabetes might be attributed to the morbidity of kidney cancer [41][42][43][44]. Currently, the prevalence of both conventional and ambulatory hypertension correlated inversely with national HDI [45]. The global burden of obesity and diabetes was also predominantly driven by high-income countries as well [46]. As with bladder cancer, smoking is one of the most essential reasons for kidney cancer [47].
Furthermore, the frequency and quality of cross-sectional imaging tend to be higher in developed nations [22]. Similarly, MIR of kidney cancer negatively correlated with certain national HDI. The stage of kidney cancer at first diagnosis crucially determines its prognosis and differs markedly across countries. Smaller masses tend to be identified easier at presentation in developed countries for non-specific symptoms [47]. Same as immunotherapy, targeted therapy and advanced surgery procedures.
The global burden and MIR of kidney cancer stagnated or decreased in the majority of countries examined from 2012 to 2018. As discussed, more developed preventative efforts and treating methods could lead to a decrease in cancer incidence and mortality.
There are some limitations to our study. First, cancer registration in relatively lessdeveloped nations could suffer from higher chance of under-reporting due to limited communication infrastructure and less robust recording system; low income and lower willingness to utilize healthcare services; relative lack of clinical services and investigation tests. GLOBOCAN and CONCORD-3 often extrapolates data for certain developing nations based on data from subnational areas or major cities. Second, discrepancies between the reliability of incidence and mortality reporting limit MIR interpretation, as mortality data are generally more accurate than incidence. Although we exclude extreme values, there is no way to correct this bias in our analysis. Third, we could not establish cause-and-effect relationships in correlational analysis.

Conclusions
In conclusion, HDI values are significantly correlated with urologic cancer incidence, MIRs and survival rates. More developed countries are more likely to have higher incidence and

Conflict of interest
The authors declare no conflicts of interest.

Ethical approval
This article does not contain any studies with human participants or animals performed by any of the authors.   The association between incidence rates of urologic cancers and HDI. (a) The