Translate this page into:
Region-wise lung cancer burden, long-term trend & time-series forecasts in India: An analytical study by age, gender, & morphology
For correspondence: Prof Aleyamma Mathew, Division of Cancer Epidemiology & Biostatistics, Regional Cancer Centre, Thiruvananthapuram 695 011, Kerala, India e-mail: aleyammamathewrcc@gmail.com
-
Received: ,
Accepted: ,
Abstract
Background & objectives
Lung cancer is the most diagnosed cancer and leading cause of cancer deaths. We assessed regional patterns in incidence, mortality, morphology, and mortality-to-incidence ratio across 57 populations, along with tobacco and alcohol use in India. We also estimated time-trends (average annual percent change: AAPC) by gender and age, and forecasted to 2030.
Methods
We used age-standardized incidence rate (ASIR) and age-standardized mortality rate (ASMR) (per 105 population) estimated AAPC via joinpoint regression and applied auto-regressive integrated moving average (ARIMA) model to forecast rates.
Results
Higher incidence of lung cancer among men was observed in the south, north and north-east regions of India. Highest ASIR was in Srinagar (39.5) and highest ASMR in Aizawl (27.1). Among women, highest rates were observed in the north-east, particularly in Aizawl (ASIR:33.7, ASMR:23.2). Tobacco use among women remains low outside the north-east, correlating with the patterns of lung cancer. Mortality to incidence ratio was low (<30%) except in a few populations. An increasing-trend in incidence was noted, with the highest AAPC in Thiruvananthapuram (women:6.7) and Dindigul (men:4.3). Adenocarcinoma has emerged as the dominant subtype over 25 yr, with higher prevalence among women, especially in Bengaluru (56.0%). Large-cell carcinoma increased notably in Delhi. By 2030, ASIRs are projected to range from 1.8 (Barshi) to 33.1 (Kollam) in men, and 1.9 (Barshi) to 8.1 (Bengaluru) in women.
Interpretation & conclusion
The burden of lung cancer in India shows regional disparities, with more adenocarcinoma, especially among women. Incidence is projected to rise, while low mortality to incidence ratio suggests underreported mortality, underscoring the need for better death reporting. Region-specific research beyond tobacco use is essential.
Keywords
ARIMA
Forecasting
joinpoint regression
lung cancer incidence and mortality
time-trends
Lung cancer is the most commonly diagnosed cancer and the leading cause of cancer-related deaths. In 2022, the global age-standardized incidence rate (ASIR) per 105 was 32.1 in men and 16.2 in women, while the age-standardized mortality rate (ASMR) was 24.8 in men and 9.8 in women1. In India, the ASIR was 8.5 in men and 3.2 in women, and the ASMR was 7.8 in men and 2.9 in women in 20221. The incidence of lung cancer in India varied geographically, with higher rates in north-east populations. Among men, the highest ASIR was reported in Aizawl (38.8), and the lowest in Barshi (1.8). Among women, the highest ASIR was reported in Aizawl (37.9), and the lowest in Osmanabad and Beed (1.0) during 2012-20162. In India, the estimated mortality-to-incidence ratio (MIR) in 2022 was 91.8 per cent among men and 90.6 per cent among women1.
Tobacco smoking is the leading cause of lung cancer globally, studies across different populations have consistently shown a strong association between tobacco use and lung cancer, and it depends on start age, duration and intensity of smoking3. Tobacco prevalence was 38 per cent among men and 8.9 per cent among women in India, with notable geographical variation. In Aizawl, which had the highest incidence of lung cancer in India, tobacco prevalence was 68.7 per cent among men and 54.7 per cent among women. In Barshi, which reported the lowest incidence rate of lung cancer among men, prevalence was 39.2 per cent, while in Osmanabad and Beed, where the incidence of lung cancer was lowest among women, prevalence was 7 per cent4. Alcohol use is another risk factor, which may contribute indirectly to lung cancer, as high levels of alcohol intake often correlated with heavy smoking. Previous study reported J-shaped association between alcohol use and lung cancer5. Alcohol prevalence in India was 18.8 per cent in men and 1.3 per cent in women, and also showing geographical variation. In Aizawl, prevalence of alcohol intake was 25.8 per cent among men and 1 per cent among women, whereas in Barshi, it was 13.1 per cent among men, and in Osmanabad and Beed, it was 0.4 per cent among women4.
An increasing-trend in the incidence of lung cancer was reported among both genders across various populations in India2. The ASIR rose from 8.5 in 2003 to 10.2 in 2017 among men, and from 3.1 in 2003 to 4.4 in 2017 in women6. The burden of lung cancer is largely driven by the distribution of morphological-subtypes, including adenocarcinoma, squamous-cell carcinoma (SqCC), large-cell carcinoma (LCC), small-cell carcinoma (SCC), and others7.
Some studies have forecasted rates of lung cancer using the auto-regressive integrated moving-average (ARIMA) model and evaluated accuracy through forecast error8-9. This study aims to assess the region-wise magnitude of (i) incidence, mortality and MIR due to lung cancer, (ii) tobacco and alcohol prevalence, and (iii) lung cancer incidence pattern by morphological-subtypes. Additionally, the study estimates time-trends in lung cancer incidence by age and gender using joinpoint regression (JR) and forecast rates through 2030 in India using the ARIMA model.
Materials & Methods
This analytical study was conducted by the division of Cancer Epidemiology & Biostatistics, Regional Cancer Centre, Thiruvananthapuram, Kerala, India after obtaining the ethical clearance from the Institutional Ethics Committee. The study was conducted during January-June 2025.
Materials
Lung cancer incidence, mortality, MIR and incidence by morphology were extracted from Cancer Incidence in Five Continents (CI5) volumes10-11, CI5plus12, research articles7,13 and population-based cancer registry (PBCR) of Thiruvananthapuram and Kollam during 1978-2022. The data includes gender-wise ASIR, ASMR, and crude incidence rate (CIR, per 105) by age (<60 and ≥60 yr). The ASMR was estimated by multiplying the MIR to ASIR. Tobacco and alcohol prevalence by gender were extracted from the 5th round of National Family Health Survey (NFHS-5)4 (Table I)4,7,10-13. Populations were classified by six regions based on the zonal councils of India ( https://www.mha.gov.in/en/page/zonal-council ), and tobacco and alcohol prevalence data (2019-2021) were mapped from the nearest corresponding areas4 (State/Union Territory/District fact sheets) (Supplementary Table I).
| Objective | Population | Source | Period | Variables |
|---|---|---|---|---|
| Region-wise LC incidence, mortality and mortality-to-incidence ratio |
57 populations with 6 regions North: 9, South: 13, East: 2, West: 10, Central: 4, North-East: 19 |
Research article13& CI5 Volume XII11 | 2015-2019 | ASIR, ASMR & MIR |
| Region-wise tobacco and alcohol use |
57 populations with 6 regions North: 9, South: 13, East: 2, West: 10, Central: 4, North-East: 19 |
5th Round National Family Health Survey (NFHS-5)4 | 2019-2021 | Proportion (%) of adults (≥15 yr) use any kind of tobacco and consume alcohol in the nearest area of the population (district or State) |
| Pattern of LC incidence by morphological subtypes |
24 populations with 5 regions North: 4, South: 6, West: 6, Central: 1, North-East: 7 |
CI5 Volume XII11 & PBCR (Kollam &Thiruvananthapuram) | 2012-2019 | Proportion (%) of LC by AdC, SqCC, LCC, SCC & others |
| Changes of LC incidence by morphological subtypes |
5 populations Bangalore, Chennai, Delhi, Mumbai, Pune |
CI5 Volume VIII10 & Volume XII11 | 1993-1997 & 2013-2017 | Proportion (%) of LC by AdC, SqCC, LCC, SCC |
| Time-trends & forecasts of LC incidence |
9 populations Bangalore: 1982-2014 Barshi: 2003-2017 Bhopal: 1988-2015 Chennai: 1983-2017 Delhi: 1988-2014 DindigulAmbilikkai: 2003-2017 Kollam*: 1991-2022 Mumbai: 1978-2017 Thiruvananthapuram#: 2005-2022 |
CI5plus12, Research article7 & PBCR (Kollam and Thiruvananthapuram) | 1978-2022 | Year-wise LC by CIR (<60 & ≥60 yr) and ASIR |
*Kollam: Karunagappally taluk, #Thiruvananthapuram: Thiruvananthapuram taluk. LC, lung cancer; CIR, crude incidence rate; ASIR, age-standardized incidence rate; ASMR, age-standardized mortality rate, Rates per 105 populations; MIR, mortality-to-incidence ratio; AdC, adenocarcinoma; SqCC, squamous-cell carcinoma; LCC, large-cell carcinoma; SCC, small-cell carcinoma; Others, other and unspecified morphology of lung cancer; CI5,cancer incidence in five continents; PBCR, population-based cancer registry
Time-trends
We used a piecewise-linear regression model with varying slopes to estimate time-trends in lung cancer incidence across nine populations in India, by gender and age (<60 and ≥60 yr) where long-term data were available. To capture non-monotonic patterns, we incorporated multiple trend segments using JR model. We applied segmented regression models with 0 to 5 joinpoints on the logarithmic scale to model trends in ASIR and CIR, under the assumption of uncorrelated errors with constant variance. Permutation test was used to select the optimal model14. The annual percent change (APC) was used to estimate the average trends within each line segment, while the average annual percent change (AAPC) was used to determine overall trends. AAPC was calculated as a weighted average of the slope of APCs from the optimal JR model, with weights corresponding to the length of each APC interval. The 95% confidence intervals (CI) for the AAPC was computed using empirical quantile method15.
We assessed the linearity assumption within each line segment graphically. The goodness of fit of the optimum JR model was evaluated using the mean squared error, calculated as the ratio of the sum of squared errors to the degrees of freedom [n-2*(r + 1)], where n is the number of observations and r is the number of joinpoints. Time-trend analyses were performed by Joinpoint Regression Program16.
Forecasting
ARIMA model was used to forecasts the incidence of lung cancer in nine populations by gender in India, denoted as ARIMA (p, d, q), where p-order of the auto-regressive component, d-degree of differencing, and q-order of the moving-average component. Leave-future-out cross-validation principle was used to estimate the forecast error, and measured the accuracy using normalised root mean squared error, where the mean of the observed data was the normalising factor17 (Supplementary material I). We estimated the coefficients of the optimum ARIMA modeland forecasted incidence rates along with their 95% prediction intervals (PI) up to 2030 using entire dataset.
Akaike information criterion [2*k-2*ln(L̂)] and Bayesian information criterion [k*ln(n)-2*ln(L̂)] were used to evaluate the goodness of fit of the optimum ARIMA model, where n is the number of observations, k is the number of parameters in the model and L̂ is the maximised value of the log-likelihood function of the model. Local polynomial smoothing was applied to visualize the values in the plots18. Time-series analyses were performed using R, and ARIMA models were fitted by Arima function from the forecast package19.
Results
Region-wise lung cancer incidence, mortality, MIR, tobacco and alcohol prevalence
Among men in the northern region, the highest ASIR was reported Srinagar (39.5), followed by Pulwama (27.8). Tobacco use prevalence in these areas was 36.5 per cent and 41.2 per cent, respectively, while alcohol use was negligible. Among women, the highest ASIR was observed in Pulwama (9.5), followed by Srinagar (9.3), with generally low tobacco and alcohol use in these regions. Chandigarh reported the highest ASMR and the highest MIR was seen in SAS Nagar and Chandigarh in both genders (Figs. 1 and 2).
- Burden of lung cancer (Incidence, mortality, mortality-to-incidence ratio), and tobacco and alcohol prevalence among men in India.
- Burden of lung cancer (Incidence, mortality, mortality-to-incidence ratio), and tobacco and alcohol prevalence among women in India.
In the southern region, Kannur reported the highest ASIR among men (35.4), followed by Kasargod (26.6), with tobacco and alcohol prevalence below 15 per cent in both areas. Dindigul reported the highest tobacco prevalence (27.7%), despite a low ASIR (5.6). Visakhapatnam had the highest alcohol prevalence (30.2%), with an ASIR of 5.2. Kollam reported the highest ASMR (20.3) and MIR. Among women, the highest ASIR was in Hyderabad (6.8), followed by Bangalore (6.2). Tobacco use was minimal (Dindigul: 6.5%; Visakhapatnam:6.3%) and alcohol use remained below 2 per cent across all populations (Figs. 1 and 2).
In the eastern region, Kolkata reported the highest ASIR of 20.4 among men and 6.6 among women, with corresponding ASMRs of 11.5 and 3.6, respectively. Tobacco and alcohol prevalence among men were 42.4 per cent and 22.2 per cent, and among women, 6 and 1 per cent, respectively. In Muzaffarpur, tobacco prevalence was 51.5 per cent among men and 7.2 per cent among women, despite low ASIRs. The MIR in Kolkata was ⁓55 per cent in both genders (Figs. 1 and 2).
In the western region, among men, the highest ASIR in Mumbai (9.8), with tobacco and alcohol prevalence at 20.3 per cent and 16 per cent, respectively. Wardha had the highest tobacco (52.6%) and alcohol (24.6%) prevalence but a low ASIR (5.1). Tobacco use in Osmanabad and Beed, Nagpur and Barshi was also high (∼40.0%), yet ASIRs remained low. Among women, the highest ASIR was also in Mumbai (5.3), with tobacco and alcohol prevalence at 5.3 per cent and 1.1 per cent, respectively. Ratnagiri reported the highest tobacco use among women (18.4%), though ASIR was only 1.8. Alcohol use among women was negligible across the region. Mumbai had the highest ASMRs and the MIR was highest in Barshi and Wardha in both genders (Figs. 1 and 2).
In the central region, Gautam Buddha Nagar reported the highest ASIR among men (16.9), with tobacco and alcohol prevalence at 29.7 per cent and 17.3 per cent, respectively. Prayagraj reported highest tobacco prevalence (44.6%), but had a low ASIR (5.8). Among women, the highest ASIR was also in Gautam Buddha Nagar (6.3). Tobacco prevalence in Prayagraj was 11.9 per cent, but the AISR remained low (2.3). The highest ASMR in Bhopal and the highest MIR in Varanasi in both genders (Figs. 1 and 2).
In the north-eastern region, Aizawl reported the highest ASIR (men: 35.9; women:33.7). Tobacco prevalence was 68.7 per cent in men and 54.7 per cent in women, while alcohol prevalence was 25.8 per cent and 1 per cent, respectively. Tobacco use exceeded 40 per cent in all these populations, and alcohol use exceeded 30 per cent in most, with the highest reported in Dima Hasao (57.2%). Among women, alcohol prevalence peaked in West Arunachal (25.0%). The highest ASMR was also in Aizawl (men:27.1; women:23.2). The MIR was highest in Aizawl and East Kashi Hills in both genders (Figs. 1 and 2).
Region-wise lung cancer incidence by morphological subtypes
Among men in 24 populations, adenocarcinoma was the most common in 15 populations, with the highest prevalence in Bengaluru (47.7%), followed by Wardha (46.4%). SqCC predominated in north-eastern region, notably in West-Arunachal (38.3%). LCC was most common in Delhi (26.7%). Among women, adenocarcinoma was the most common subtype in 20 populations, with the highest in Ahmedabad (59.0%), followed by Bengaluru (55.7%). SqCC predominated in north-eastern region, notably in West-Arunachal (34.2%). In Delhi, LCC was accounted for 25.3% (Table II).
| Population | Men | Women | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| AdC | SqCC | LCC | SCC | Others | AdC | SqCC | LCC | SCC | Others | |
| North | ||||||||||
| Chandigarh | 22.1 | 20.4 | 4.3 | 15.4 | 37.9 | 39.3 | 7.1 | 4.8 | 14.3 | 34.5 |
| Delhi | 17.1 | 16.1 | 26.7 | 9.6 | 30.4 | 31.1 | 7.6 | 25.3 | 6.3 | 29.8 |
| Sangrur | 14.1 | 19.6 | 9.8 | 10.9 | 45.7 | 36.2 | 4.3 | 6.4 | 2.1 | 51.1 |
| SAS Nagar | 20.0 | 27.6 | 3.8 | 13.0 | 35.7 | 31.4 | 7.8 | 0.0 | 7.8 | 52.9 |
| South | ||||||||||
| Bengaluru | 47.4 | 15.0 | 12.2 | 5.1 | 20.2 | 55.7 | 11.0 | 12.7 | 4.1 | 16.5 |
| Chennai | 23.0 | 13.8 | 14.4 | 3.9 | 44.9 | 32.5 | 7.6 | 14.4 | 0.9 | 44.7 |
| DindigulAmbilikkai | 32.6 | 13.0 | 15.3 | 2.9 | 36.3 | 37.3 | 7.6 | 8.5 | 5.1 | 41.5 |
| Kollam | 15.3 | 11.5 | 2.8 | 6.5 | 63.8 | 25.9 | 4.4 | 2.7 | 2.4 | 64.6 |
| Tamil Nadu | 30.5 | 16.4 | 13.8 | 4.9 | 34.4 | 41.4 | 8.7 | 13.1 | 1.6 | 35.2 |
| Thiruvananthapuram | 20.6 | 11.9 | 6.7 | 5.3 | 55.3 | 44.8 | 5.4 | 6.9 | 0.4 | 42.5 |
| West | ||||||||||
| Ahmedabad Urban | 40.3 | 22.6 | 2.6 | 8.7 | 25.8 | 59.0 | 10.0 | 4.4 | 1.6 | 25.1 |
| Aurangabad | 8.1 | 13.2 | 0.9 | 2.6 | 75.3 | 6.5 | 18.3 | 0.0 | 0.0 | 75.3 |
| Barshi Rural | 28.1 | 3.1 | 6.3 | 0.0 | 62.5 | 22.2 | 3.7 | 3.7 | 7.4 | 63.0 |
| Mumbai | 33.9 | 11.5 | 8.5 | 9.3 | 36.8 | 43.4 | 4.4 | 8.1 | 4.6 | 39.4 |
| Pune | 39.8 | 13.1 | 2.3 | 8.1 | 36.7 | 50.4 | 4.6 | 2.5 | 8.6 | 33.8 |
| Wardha | 46.4 | 10.5 | 5.0 | 12.7 | 25.4 | 52.8 | 5.6 | 2.8 | 5.6 | 33.3 |
| Central | ||||||||||
| Bhopal | 16.8 | 19.1 | 1.3 | 8.3 | 54.5 | 16.5 | 22.4 | 0.0 | 8.2 | 52.9 |
| North-East | ||||||||||
| Dibrugarh | 24.8 | 15.8 | 2.3 | 4.5 | 52.6 | 25.5 | 10.9 | 0.0 | 1.8 | 61.8 |
| Kamrup Urban | 33.0 | 14.7 | 2.7 | 2.9 | 46.7 | 38.5 | 6.8 | 1.4 | 2.7 | 50.7 |
| Manipur | 20.1 | 31.0 | 2.8 | 6.2 | 40.0 | 24.2 | 24.8 | 3.6 | 8.1 | 39.3 |
| Meghalaya | 19.2 | 27.2 | 10.2 | 7.4 | 35.9 | 32.7 | 14.2 | 8.0 | 4.4 | 40.7 |
| Mizoram | 10.4 | 20.4 | 7.1 | 7.8 | 54.3 | 11.0 | 15.2 | 5.6 | 9.4 | 58.8 |
| Tripura | 35.7 | 18.0 | 6.1 | 9.3 | 30.9 | 44.1 | 13.2 | 6.4 | 7.3 | 29.1 |
| West Arunachal | 30.0 | 38.3 | 11.7 | 3.3 | 16.7 | 36.8 | 34.2 | 7.9 | 0.0 | 21.1 |
Others, others and unspecified morphology of lung cancer
Based on data from five longstanding PBCRs in India (1993-97 to 2013-17), among men, the proportion of adenocarcinoma increased, most notably in Bengaluru (13.1% to 47.4%). LCC also increased, with the highest increase observed in Delhi (1.3% to 26.7%). SCC remained relatively stable. SqCC decreased overall, with the steepest decline in Mumbai (21.8% to 11.5%). Among women, adenocarcinoma increased as well, most notably in Bengaluru (23.9% to 55.7%). LCC also rise, with the highest in Delhi (1.0% to 25.3%). SCC remained relatively stable. SqCC decreased in most regions, except for an increase in Bengaluru (7.1% to 11.0%) (Fig. 3).
- Changes of lung cancer incidence (%) by Morphological subtypes in India: 1993-1997 to 2013-2017 in (A) men and (B) women. AdC, adenocarcinoma; LCC,large-cell carcinoma; SqCC, squamous-cell carcinoma; SCC,small-cell carcinoma. Delhi: 1993-1996 to 2013-2015, Bengaluru: 1993-1997 to 2013-2015.
Time-trends in lung cancer incidence
Among men of all ages combined, an increasing-trend in incidence was observed in seven out of nine populations with the highest AAPC in Dindigul (4.3), followed by Thiruvananthapuram (2.9) and Bengaluru (2.8). In contrast, a decreasing-trend was noted in Mumbai (-0.8). Upward trends were also observed in both <60 and ≥60 yr. Among women of all ages combined, an increasing trend in incidence was observed in 7 out of 8 populations, with the highest AAPC in Thiruvananthapuram (6.7), followed by Dindigul (5.5) and Kollam (5.3). Similarly increasing-trend were observed in both <60 and ≥60 yr (Supplementary Tables II-IV).
Forecast in lung cancer incidence
The forecast error and optimal model summary for the nine populations are provided in supplementary tables V-VI. Among men, the forecasted ASIR for the year 2030 along with 95%PI are as follows; Bengaluru:15.3 (4.2-26.5), Barshi:1.8 (0.01-3.7), Bhopal:12.1 (9.5-14.8), Chennai:14.8 (9.1-20.5), Delhi:20.5 (11.1-29.9), Dindigul:8.5 (7.2-9.9), Kollam:33.1(28.6-37.6), Mumbai:8.7 (3.8-13.7) and Thiruvananthapuram:24.4 (20.8-28.1). Among women, the corresponding forecasted ASIRs (95%PI) for 2030 are; Bengaluru: 8.1 (4.9-11.3), Barshi: 1.9 (0.01-3.9), Chennai: 6.2 (5.1-7.3), Delhi: 6.9 (5.3-8.5), Dindigul: 2.8 (2.3-3.4), Kollam: 7.6 (5.4-9.8), Mumbai: 6.1 (1.8-10.3) and Thiruvananthapuram: 7.6 (4.2-10.9) (Fig. 4).
- Lung cancer incidence by gender in India: 1978-2030 in (A) men and (B) women.
Discussion
This is the first and most up-to-date analysis of burden of lung cancer and tobacco and alcohol use across 57 populations in six Indian regions. We identified regional disparities in incidence rates: higher lung cancer rates among men in the south, north, and north-east and among women in the north-east. Previous study linked high lung cancer rates among both men and women in the north-east to elevated tobacco use7. Women-to-men incidence and mortality ratios in the north-east nearing 1 further suggest similar tobacco use across genders4. In contrast, southern region reported lower substance use (<25%)4 but still reported high lung cancer ASIR rates, indicating other contributing factors.
In the present analysis, ASMR and MIR varied across India, but were consistent between genders within each population. Globally, lung cancer shows high MIR, even in developed countries20. While World Health Organization estimated India’s MIR above 90 per cent in 20221, the present study found such levels only in three PBCRs. Additionally, 45 per cent of lung cancer cases in India are diagnosed at a distant stage2, contributing to high mortality. Low MIR may reflect underreporting of lung cancer deaths, highlighting the need to strengthen cause-of-death data.
An increasing-trend in the incidence of lung cancer was observed among men, consistent with other studies2,7. The highest ASIRs over the time was observed in Kollam, where tobacco use reached 56.8 per cent21. Despite declining tobacco use4, current lung cancer burden likely reflects past exposure due to its long latency (10–30 yr)22-23. Women showed a higher AAPC than men. Globally24 and in India7, lung cancer is rising among women, including younger age groups, despite tobacco use being <10 per cent4, suggesting other contributing factors. Lung cancer incidence was also higher and increasing among older adults in both genders, likely due to aging and prolonged exposure to tobacco and alcohol, which are more prevalent in older men4,25.
Adenocarcinoma was the most common and fastest-growing lung cancer subtype, especially among women, consistent with global trends7,26. Around 25 per cent of lung cancer cases worldwide are linked to non-tobacco factors27. A study identified particulate matter in air pollution as a major cause of lung cancer28. In India, an estimated 1.7 million deaths are attributable to air pollution29. Biomass fuel use30 and second-hand smoke exposure31 (reported in 48.5% of adults32), are also key contributors to lung cancer, particularly among women.
This study used NFHS data to examine regional lung cancer patterns and key risk factors like tobacco and alcohol. Other factors, such as air pollution and occupational exposure, were excluded due to lack of population-specific data. Additionally, limited longitudinal data hindered analysis of exposure–disease latency. Some populations showed inconsistency between lung cancer rates and tobacco use, particularly among men with high tobacco use but low lung cancer rates, possibly due to underreporting or patients seeking care elsewhere. In Mumbai, a sharp decline in male lung cancer incidence, also noted in other studies2,7, may reflect data collection issues.
While overall lung cancer trends were non-linear, they were linear within each joinpoint, supporting the use of the JR model. However, limited data in certain populations and younger age groups led to unstable estimates. Additionally, a high proportion of unspecified morphologies limited subtype-specific and mortality trend analyses.
In conclusion, regional disparities in lung cancer burden were observed, with a higher proportion of adenocarcinoma, particularly among women. An upward trend in incidence is projected in the coming decades. Low MIR suggests underreported mortality, highlighting the need for improved death reporting. Strengthening data systems, diagnostics, and targeted prevention is crucial. Further research is needed to identify region-specific lung cancer drivers beyond tobacco, including environmental, occupational, and lifestyle factors, key to effective prevention and early detection.
Acknowledgment
Authors acknowledge International Agency for Research on Cancer and National Cancer Registry Programme for accessing the lung cancer incidence and mortality data of India.
Financial support & sponsorship
None.
Conflicts of Interest
None.
Use of Artificial Intelligence (AI)-Assisted Technology for manuscript preparation
The authors confirm that they used ChatGPT ( https://chat.openai.com/chat ) for improving language and readability at the initial draft sections and no images were manipulated using AI.
References
- International Agency for Research on Cancer. World Health Organization. Global cancer observatory. Cancer today. Fact sheets. Available from: https://gco.iarc.who.int/today, accessed on March 3, 2025.
- Indian Council of Medical Research-National Centre for Disease Informatics and Research. Report of National Cancer Registry Programme 2020. Available from: https://www.ncdirindia.org/All_Reports/Report_2020/default.aspx, accessed on May 2, 2025
- Etiology of lung cancer: Evidence from epidemiologic studies. J Natl Cancer Cent. 2022;2:216-25.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- The DHS Program. Demographic and Health Suveys. Publications Summary. National, State and Union Territory, and District fact sheets 2019-21, National Family Health Survey (NFHS-5). Available from: https://dhsprogram.com/publications/publication-OF43-Other-Fact-Sheets.cfm, accessed on 10 May 2025
- Alcohol consumption and lung cancer risk: A pooled analysis from the International Lung Cancer Consortium and the SYNERGY STUDY. Cancer Epidemiol. 2019;58:25-32.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- International agency for research on cancer. World Health Organization. Global cancer observatory: Cancer over time. Available from: https://gco.iarc.who.int/overtime, accessed on March 5, 2025.
- A clinicoepidemiological profile of lung cancers in India - Results from the national cancer registry programme. Indian J Med Res. 2022;155:264-72.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Predicting the burden of cancer in Switzerland up to 2025. PLOS Glob Public Health. 2022;2:e0001112.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Comparison of statistical models to predict age‐standardized cancer incidence in Switzerland. Biometrical J. 2023;65:e2200046.
- [Google Scholar]
- Parkin DM, Whelan SL, Ferlay J, Teppo L, Thomas DB, eds. Cancer Incidence in Five Continents, Vol. VIII.IARC Scientific Publication No. 155. Lyon: International Agency for Research on Cancer; 2002.
- Bray F, Colombet M, Aitken JF, Bardot A, Eser S, Galceran J, eds. Cancer Incidence in Five Continents, Vol. XII.IARC Scientific Publication No. 169. Lyon: International Agency for Research on Cancer; 2024.
- Cancer incidence in five continents, CI5plus: IARC Cancer Base No. 9. Lyon: International agency for research on cancer;
- Cancer incidence and mortality across 43 cancer registries in India. JAMA Netw Open. 2025;8:e2527805.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Permutation tests for joinpoint regression with applications to cancer rates. Stat Med. 2000;19:335-51.
- [CrossRef] [PubMed] [Google Scholar]
- Twenty years since joinpoint 1.0: Two major enhancements, their justification, and impact. Stat Med. 2022;41:3102-30.
- [CrossRef] [PubMed] [Google Scholar]
- National Cancer Institute. Division of Cancer Control & Population Sciences. Surveillance Research Program. Joinpoint Trend analysis software. Available from: https://surveillance.cancer.gov/joinpoint/, accessed on May 20, 2025
- Approximate leave-future-out cross-validation for Bayesian time series models. J Stat Computation Simulation. 2020;90:2499-523.
- [Google Scholar]
- Local regression models. In: Chambers JM, Hastie TJ, eds. Statistical models in S. Routledge; 2017. p. :309-76.
- [Google Scholar]
- R Core Team. R: A language and environment for statistical computing (Version 4.3.1). R Foundation for Statistical Computing, Vienna, Austria 2023. Available from: https://www.R-project.org/, accessed on May 25, 2025.
- Mapping of global, regional and national incidence, mortality and mortality-to-incidence ratio of lung cancer in 2020 and 2050. Int J Clin Oncol. 2022;27:665-7.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Prevalence and determinants of tobacco habit in Kollam district, Kerala, India. Tampere University Press; 2014.
- Longitudinal relationships between lung cancer mortality rates, smoking, and ambient air quality: a comprehensive review and analysis. Crit Rev Toxicol. 2019;49:790-818.
- [CrossRef] [PubMed] [Google Scholar]
- Exposure–lag response of smoking prevalence on lung cancer incidence using a distributed lag non-linear model. Sci Rep. 2021;11:14478.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Lung cancer incidence in young women vs. young men: a systematic analysis in 40 countries. Int J Cancer. 2020;147:811-9.
- [Google Scholar]
- Cancer treatment and survivorship statistics, 2022. CA Cancer J Clin. 2022;72:409-36.
- [CrossRef] [PubMed] [Google Scholar]
- Global variations in lung cancer incidence by histological subtype in 2020: a population-based study. Lancet Oncol. 2023;24:1206-18.
- [CrossRef] [PubMed] [Google Scholar]
- Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74:229-63.
- [CrossRef] [PubMed] [Google Scholar]
- Air pollution and lung cancer incidence in 17 European cohorts: Prospective analyses from the European study of cohorts for air pollution effects (ESCAPE) Lancet Oncol. 2013;14:813-22.
- [CrossRef] [PubMed] [Google Scholar]
- Health and economic impact of air pollution in the states of India: The global burden of disease study2019. Lancet Planet Health. 2021;5:e25-38.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Does household use of biomass fuel cause lung cancer? A systematic review and evaluation of the evidence for the GBD 2010 study. Thorax. 2015;70:433-41.
- [CrossRef] [PubMed] [Google Scholar]
- Association between second-hand smoke exposure and lung cancer risk in never-smokers: a systematic review and meta-analysis. Eur Respir Rev. 2024;33:240077.
- [CrossRef] [PubMed] [PubMed Central] [Google Scholar]
- Indian Council of Medical Research-National Centre for Disease Informatics and Research. National Noncommunicable Disease Monitoring Survey (NNMS) 2017-18. Available from: https://www.ncdirindia.org/nnms/, accessed on May 5, 2025.