Article Text
PDF
Abstract
Introduction Lung cancer has a poor prognosis that varies internationally when assessed by the two major histological subgroups (non-small cell (NSCLC) and small cell (SCLC)).
Method 236 114 NSCLC and 43 167 SCLC cases diagnosed during 2010–2014 in Australia, Canada, Denmark, Ireland, New Zealand, Norway and the UK were included in the analyses. One-year and 3-year age-standardised net survival (NS) was estimated by sex, histological type, stage and country.
Results One-year and 3-year NS was consistently higher for Canada and Norway, and lower for the UK, New Zealand and Ireland, irrespective of stage at diagnosis. Three-year NS for NSCLC ranged from 19.7% for the UK to 27.1% for Canada for men and was consistently higher for women (25.3% in the UK; 35.0% in Canada) partly because men were diagnosed at more advanced stages. International differences in survival for NSCLC were largest for regional stage and smallest at the advanced stage. For SCLC, 3-year NS also showed a clear female advantage with the highest being for Canada (13.8% for women; 9.1% for men) and Norway (12.8% for women; 9.7% for men).
Conclusion Distribution of stage at diagnosis among lung cancer cases differed by sex, histological subtype and country, which may partly explain observed survival differences. Yet, survival differences were also observed within stages, suggesting that quality of treatment, healthcare system factors and prevalence of comorbid conditions may also influence survival. Other possible explanations include differences in data collection practice, as well as differences in histological verification, staging and coding across jurisdictions.
- Histology/Cytology
- Lung Cancer
- Non-Small Cell Lung Cancer
- Small Cell Lung Cancer
Data availability statement
Data may be obtained from a third party and are not publicly available. Data contain sensitive and personal information and can be obtained from the cancer registries listed as the local leads in the manuscript.
Statistics from Altmetric.com
Key messages
What is the key question?
Are there international disparities in lung cancer survival by clinically relevant subgroups in the most recent population-level data?
What is the bottom line?
International differences in survival among patients with lung cancer persist in high-income countries, which are partly due to differences in stage at diagnosis or early detection, and to survival difference within stage groups due to cancer management.
Why read on?
This study presents in-depth results of the most up-to-date differences of lung cancer stage distribution and survival by histological types, age group and sex for each included country, as well as within countries, followed by a discussion of potential causes of the disparities including clinical and data factors.
Introduction
Lung cancer is the leading cause of cancer death among men and women worldwide, with an estimated 1.6 million deaths, or nearly 20% of all cancer deaths, occurring in 2018.1 Lung cancer is categorised into two main histological groups—small cell lung carcinoma (SCLC) and non-small cell lung carcinoma (NSCLC)—with patients with SCLC generally having poorer outcomes compared with NSCLC.2 Marked sex differences in survival have also been previously reported, with women having a more favourable prognosis.3–6
As cancer survival is one of the key measures of performance improvement in healthcare, international cancer survival comparisons can identify possible causes for observed variations, such as access to early detection or optimal treatments. The International Cancer Benchmarking Partnership (ICBP), a consortium of clinicians, policymakers, researchers and cancer data experts, has previously described lung cancer survival differences across high-income countries7 and concluded that stage at diagnosis and histological type explained some of the international variation in lung cancer survival in 2004–2007.2 In this study, we extend this prior work by assessing the most up-to-date lung cancer survival statistics by sex and stage at diagnosis for NSCLC and SCLC, using population-based data from seven countries.
Methods
Data
Data were collected as part of the ICBP SURVMARK-2 Project for patients diagnosed during 1995–2014 at ages above 18 years and followed until the end of 2015.8 ICBP countries were included due to their high-quality population-wide registries, universal access to healthcare with similar healthcare expenditure, and interest in understanding how and why cancer survival differs across countries. For this study, lung carcinoma data from 18 population-based cancer registries in seven countries with at least 50% complete data on stage at diagnosis during 2010–2014 were included (Australia—New South Wales; Canada—Alberta, British Columbia, Manitoba, New Brunswick, Newfoundland, Nova Scotia, Ontario, Prince Edward Island, Saskatchewan; Denmark; Ireland (2010–2013); New Zealand; Norway; and the UK—England, Scotland, Wales, Northern Ireland). Cases diagnosed with an invasive primary malignant tumour of the lung and bronchus (International Classification of Diseases 10th Revision: C34) were classified as SCLC, NSCLC, other and unspecified histology (see table 1). Of 393 453 adult patients with lung cancer diagnosed during 2010–2014, we excluded cases diagnosed based on death certificate only or at autopsy (N=8105, 2.1%), with inconsistent dates (N=54, <0.0%), below age 15 years or above age 99 years (N=127, <0.0%), with second primary cancer in the lung (N=5034, 1.3%), with incorrect stage information (N=4557, 1.2%) or classified as neuroendocrine neoplasms (N=7869, 2.0%).
Characteristics of patients diagnosed with lung cancer during 2010–2014
Each participating cancer registry provided information on pretreatment pathological and clinical stage based on the tumour extent (T), the degree of nodal involvement (N) and the presence of metastases (M), grouped TNM (stages I–IV), and SEER summary stage 2000 (online supplemental figure 1). A previously developed algorithm9 was adapted to harmonise individual or grouped TNM information (seventh edition10) to SEER stage (localised, regional, distant and missing), creating a ‘mapped SEER stage’ that combines both TNM and SEER stage information; during the stage mapping, pathological T and N, and clinical M information was prioritised if both pathological and clinical data were available (online supplemental figure 1).
Supplemental material
Statistical analyses
We estimated net survival, which is the survival of patients with cancer after adjustments for background mortality in the general population using life tables. Life tables were specific to sex, single year of age, geographical region and year of death. Net survival estimates at 1 and 3 years after diagnosis were estimated by histological group and sex for each country and jurisdiction, as appropriate, using Pohar Perme estimators.11 Age standardisation was carried out using International Cancer Survival Standard weights.12 The period approach was used to provide short-term predictions of 3-year survival for patients diagnosed during 2010–2014. Statistically significant differences in net survival between sexes, histological subgroups, countries or jurisdictions were determined by comparing the 95% CIs.
Stage at diagnosis was imputed separately for patients with SCLC and NSCLC with missing stage data (SCLC missingness range: 1.1% (Denmark–SEER) to 33.6% (UK–TNM); NSCLC missing range: 1.1% (Denmark–SEER) to 33.3% (UK–TNM). Covariates in the imputation model included age, sex, year of diagnosis, survival time and the Nelson-Aalen estimator of the cumulative hazard. We ran the imputation model 30 times and combined the results using Rubin’s rules.13 Stage-specific survival estimates using the imputed datasets were then compared with the corresponding results without imputation (online supplemental tables 1 and 2). All analyses were undertaken using Stata V.14.14
Final results are presented by grouped TNM for all countries except Australia and New Zealand where TNM data were not collected, and mapped SEER stage for all countries, and stratified by histology group. All results relate to imputed stage, unless otherwise noted. For simplicity, we used stages I–IV when referring to TNM stage, and ‘localised’, ‘regional’ and ‘distant’ when referring to mapped SEER stage. Finally, while we focused on sex-specific estimates, the findings for both sexes combined are presented in online supplemental figures 2–5 and online supplemental figure 1.
Results
In total, 367 707 patients (93.5%) diagnosed during 2010–2014 were included in the survival analyses (table 1). A total of 280 744 cases (76.4% of eligible cases) were microscopically confirmed, either through cytological examination, histology of primary tumour or of metastasis. The proportion with NSCLC ranged between 59% (UK) and 76% (Australia), while the proportion with SCLC was largely consistent across all countries at 11%–12%, with a slightly larger proportion in Denmark (14%) and Norway (16%). The proportion with unspecified cancers ranged from 11% (Denmark) to 29% (UK). For the remainder of this report, we will comment only on results for NSCLC and SCLC cases (n=279 281).
Non-small cell lung carcinoma
At least 45% of the 236 114 NSCLC cases were diagnosed at the most advanced stage (stage IV for TNM, distant stage for SEER) for men in all countries, with the largest proportion observed in New Zealand at 59.8% (table 2, figure 1A). Compared with men, women had a more favourable stage distribution in all countries, except New Zealand and Denmark (table 3, figure 1A).
Number and proportion of male patients with non-small cell lung carcinoma (NSCLC) and small cell lung carcinoma (SCLC) diagnosed during 2010–2014 by country and stage at diagnosis (TNM and SEER), before and after imputation
Number and proportion of female patients with non-small cell lung carcinoma (NSCLC) and small cell lung carcinoma (SCLC) diagnosed during 2010–2014 by country and stage at diagnosis (TNM and SEER), before and after imputation
Distribution of (imputed) stage at diagnosis of (A) non-small cell lung carcinoma and (B) small cell lung carcinoma by sex and country, 2010–2014. TNM, tumour/node/metastasis.
For all stages combined, the age-standardised 1-year net survival for NSCLC among men was highest for Canada (48.1%) and lowest for Ireland (41.8%), New Zealand (41.5%) and the UK (40.8%) (figure 2A, online supplemental table 11). The 3-year net survival showed a similar pattern (between 19.7% and 27.1%). For women, the range was slightly wider for 1-year and 3-year net survival (between 48.3% and 57.8%, and 25.3% and 35.0%, respectively) (figure 2A, online supplemental table 12).
One-year and 3-year age-standardised* net survival of (A) non-small cell lung carcinoma and (B) small cell lung carcinoma by sex and country, 2010–2014. TNM, tumour/node/metastasis.
Sex-specific net survival varied substantially depending on stage at diagnosis, with little consistency in the ranking of countries at both 1 year and 3 years after diagnosis (figure 3A, online supplemental tables 11 and 12). For example, for men, Australia had the lowest 1-year net survival for localised tumours (81.6%), but the highest 1-year net survival for distant tumours (25.6%). Nevertheless, we observed that survival for men and women was consistently high across all (SEER) stages for Canada, while Ireland and the UK reported consistently lower survival.
One-year and 3-year age-standardised* net survival of (A) non-small cell lung carcinoma and (B) small cell lung carcinoma by (imputed) SEER stage, sex and country, 2010–2014.
Among men and women, both 1-year and 3-year survival for distant SEER stage was highest for Australia and lowest for the UK, respectively. Further, as seen for overall NSCLC, women had higher 1-year and 3-year net survival for all stages at diagnosis. Notable sex differences (ie, >10% point difference) were observed at 1 year post-diagnosis for regional NSCLC in New Zealand (Δ=10.7% point), Ireland (Δ=12.0% point) and Norway (Δ=11.4% point).
Similar differences in stage-specific net survival were observed when using TNM stage (figure 4A, online supplemental tables 11 and 12). Among the five countries that provided TNM stage, 1-year net survival for NSCLC with TNM stages II and III was significantly lower for men in Ireland, at 69.8% and 45.3%, respectively, and the UK, at 76.3% (TNM stage II only), compared with Denmark (84.4% for stage II, 56.0% for stage III). Among cases diagnosed with stage IV disease, 1-year net survival was lowest at 18.8% for men in the UK, followed by Ireland (19.8%), Denmark (21.2%), Norway (21.8%), and finally Canada (23.6%). As for women, the lowest survival for all stages at both 1 year and 3 years post-diagnosis was observed for the UK and Ireland. The largest absolute differences in net survival between countries were observed for stages II and III, and there was an indication that survival was more homogeneous for stage IV, particularly for 3-year net survival where the absolute difference between Canada (highest) and the UK (lowest) was only 3.7% points.
One-year and 3-year age-standardised* net survival of (A) non-small cell lung carcinoma and (B) small cell lung carcinoma by (imputed) TNM stage, sex and country, 2010–2014. TNM, tumour/node/metastasis.
Small cell lung carcinoma
Approximately 70% of patients with SCLC were diagnosed with stage IV or distant disease (Tables 2 and 3, figure 1B), although some variation was observed between countries and sexes; for example, New Zealand had the largest proportion of patients diagnosed with distant disease at 84.3% for men and 78.4% for women. Women had a more favourable stage distribution compared with men.
One-year net survival ranged for men from 25.0% in New Zealand and the UK to 31.8% in Australia, and for women 31.2% in the UK to 39.2% in Canada (figure 2B, online supplemental tables 13 and 14). Canada and Norway had the highest 3-year net survival for both men and women, though differences between the seven countries were minimal. There were large disparities in net survival between sexes, in particular at 1 year after diagnosis, with 1-year net survival for women being 6.2% points (UK) to 10.6% points (Canada) higher than that observed for men.
Differences by SEER stage in net survival between the seven countries were also found (figure 3B, online supplemental tables 13 and 14). Among men with localised SCLC, 1-year net survival was lower in Australia (58.3%) and Norway (58.6%) than in Canada (82.2%) and Denmark (79.5%). For cases with regional and distant disease, men in Denmark had the highest 1-year net survival at 63.0% and 24.3%, respectively; whereas for regional disease, 1-year survival was lowest for men in Ireland and New Zealand, and Norway and New Zealand had the lowest survival for distant cases. Net survival at 3 years among male patients showed similar patterns. For women, 1-year and 3-year net survival was highest in Denmark for most stages. New Zealand and Norway had lower survival for most stages along with Ireland and the UK, though most notable was the low survival observed for women with localised stage in Australia (62.8%) at 1 year post-diagnosis (online supplemental table 14).
Net survival by TNM stage is shown in figure 4B (online supplemental tables 13 and 14). In general, similar differences in stage-specific net survival were observed when using TNM stage. For example, for women, 1-year net survival for stage III SCLC ranged from 49.1% in Ireland to 56.3% in Canada; the corresponding 3-year net survival ranged from 15.6% in the UK to 24.2% in Canada. As with NSCLC, the survival differences among the countries investigated were largest for stages II and III.
Within-country differences
While it was beyond the scope of this manuscript to explore within-country disparities for NSCLC and SCLC survival in detail, the best outcomes in Canada were generally observed for Manitoba and New Brunswick while the worst outcomes were observed for Prince Edward Island and Nova Scotia, although the differences were not always statistically significant due to low statistical power for the latter jurisdictions (online supplemental tables 6–10, 15–20). In the UK, survival for NSCLC and SCLC appeared to be highest in Scotland for early disease stages, with survival at advanced stages being generally comparable across the four jurisdictions.
Discussion
In this study, discrepancies in lung cancer survival were evident across countries by sex, histological group and stage, with higher survival observed for Canada and Norway, and lower survival in the UK, New Zealand and Ireland. Survival was consistently higher for women (for all stage and histological groups) and for NSCLC. Variations in stage distribution and stage-specific survival estimates were apparent and may partly explain international survival differences. Net survival within stage groups varied across countries whereby survival was consistently higher across stages of disease for Canada and Denmark, and consistently lower for Ireland and the UK.
A previous study of a similar group of countries reported 1-year net survival among lung cancer cases diagnosed from 2004 to 2007.2 Despite differences in study methods, generally we obtained similar results with respect to survival differences by histological group and stage, as well as the countries’ rankings. Similar findings were also noted in another study investigating lung cancer survival in Europe for 1999–2007, where Ireland and the UK generally exhibited lower survival.15 Nonetheless, a decade later, we observed improved survival for lung cancer for all countries,8 as well as higher stage-specific survival particularly for NSCLC in this study. However, the contribution of stage migration to the apparent improvement in stage-specific survival cannot be completely ruled out. For example, the migration due to coding stage III to stage IV disease resulting from increased sensitivity and utilisation of novel imaging techniques, such as positron emission tomography.
Our findings show that women generally are diagnosed at an earlier stage and have better survival outcomes than men. This is in line with previous studies3–6 evaluating the effect of sex on lung cancer prognosis and several potential reasons for this observation have been proposed. First, adenocarcinoma lung cancer is the predominant subtype of NSCLC and is more common among women than men. As adenocarcinoma lung cancer is associated with being a never smoker16 and having a better response to treatment,17 we would expect that women would experience better NSCLC survival compared with men because there is a higher proportion of these favourable cases. Second, adenocarcinoma that has a wider spectrum of growth rate with some being quite indolent is generally diagnosed at earlier stages compared with other histological types,18 which corresponds with our finding that a higher proportion of women present with early stage disease, though this may also be due to more frequent and earlier medical consultations. Finally, several studies19–21 have also suggested that female sex may be a positive prognostic factor in itself,21 22 irrespective of age, stage, period of diagnosis and histological groups. Further, mutations in epidermal growth factor receptor have been shown to be more prevalent in women, which may lead to survival advantages.23 24
Net survival for SCLC was consistently lower compared with that for NSCLC in all countries. This corresponds with the fact that SCLC tends to grow and spread faster than NSCLC, exemplified by previous research that found larger proportions of individuals with SCLC compared with NSCLC diagnosed at advanced stage or with comorbid illnesses, which may lead to fewer patients being considered for curative treatment.25 For example, in Norway, only 1.7% of patients with SCLC received surgery compared with 18.1% of those with any lung cancer in the period 2002–2011.26 While we observed considerable differences in 1-year net survival across countries for SCLC, disparities were much smaller at 3 years, which may be explained by the high fatality rate of SCLC regardless of available treatments.27
Disparities observed for NSCLC between countries may partly relate to varying proportions with squamous cell carcinoma, large cell carcinoma and adenocarcinoma, together forming the group with NSCLC. For example, for NSCLC, 50% were adenocarcinomas for Canadian men compared with 40% for men in the UK with a similar difference for women (62% vs 51%, respectively). As a result, the UK had a larger proportion of squamous cell carcinomas which are associated with lower survival overall compared with adenocarcinoma.28 Unfortunately further stratification of the NSCLC by finer histological groupings was not possible due to insufficient statistical power. Similarly, the UK and New Zealand had larger proportions with NSCLC and SCLC diagnosed at later stages compared with other countries suggesting that countries like Canada may detect cancers earlier, although a formal recommendation for lung cancer screening was made in Canada after our study period in 2016.29
Although differences in stage distribution will account for some of the survival disparities between countries, differences in stage-specific survival were also evident across countries in our study. These absolute differences were most pronounced for localised and regional stage disease, and may relate to the variability in treatment and management. For example, as surgical resection offers the only realistic chance of cure for lung cancer,30 differences in rates of surgery may partly explain survival differences in patients with localised and regional lung cancer. Differences in prehabilitation prior to lung cancer surgery or age at diagnosis across countries are other factors that may explain the observed survival differences.31 For example, the median age of patients with NSCLC in the UK was generally higher than that for other countries at each cancer stage. Previous studies have shown that patients at older ages are more likely to have comorbid conditions, which are known predictors of treatment decisions and poorer outcomes including survival.32 While historically only around 9% of patients with NSCLC in the UK underwent surgery, the current percentage is 18% according to the National Lung Cancer Audit 2018 annual report, which is above the national standard.33 Combined with the other improvements in lung cancer care in the UK, including recent adoption of immunotherapy and increases in pathological confirmation of cancer and overall cancer care,33 we expect to see improvements in stage-specific survival in the next ICBP iteration.
A strength of our study is that all data are from high-quality population-based cancer registries, which were additionally standardised and checked using a predefined protocol. All results were validated and interpreted with the input of local experts, including registry experts, epidemiologists and clinicians from each country. Yet, a number of limitations should be noted. First, there is still a substantial proportion of cases with unknown histology, ranging between 11% and 29% across the seven countries. While survival for these cases is typically lower than for cases with known histology, the size of this group might have affected survival estimates for SCLC and NSCLC. In a complementary paper, we showed that while the proportion of lung cancer cases with unknown histology has an impact on survival estimates for ‘known’ histology groups, the inclusion and/or reallocation of these individuals does not change international rankings or overall patterns we see across countries.34 Second, when mapping information from different staging systems, there is potential for misclassification, which might have affected the reported stage distributions and corresponding survival estimates. When analyses were restricted to cases with both SEER and TNM information, discrepancies in the final summary stage were found for 8% of all cases and this affected the estimated survival for localised and regional stage disease only (results not shown). Nonetheless, differences in cancer registration practices and staging systems used should be taken into account when interpreting the results of this paper.35 Finally, missing information on stage at diagnosis presents another potential concern when conducting survival comparisons by stage across countries. To some extent, we mitigated the issue of differential missingness across countries by running imputation models separately by country (or jurisdiction) and by including measures of survival time in the models. In sensitivity analyses, we showed that both approaches (with and without imputation) led to very similar results. Yet, there are factors that cannot be taken into account in multiple imputation models that may influence missingness of stage information such as differences in registration practices36 and this may have affected our results. Including information such as comorbidity or performance status may improve the imputation and also further our understanding on the causes of survival difference by stage across countries.
In conclusion, this study provides more recent estimates of stage-specific lung cancer survival for seven high-income countries with comparison for the two main histological types and investigating differences by sex. While wide international disparities in lung cancer survival persist, this study illustrates the favourable prognosis for NSCLC cases and early stage lung cancers for women. It is also evident that differences in stage distribution across sex groups, histological types and countries partly explain better lung cancer survival, yet international differences were also observed for stage-specific survival suggesting that factors linked to disease treatment and management, such as rates of high-quality surgery, utilisation of targeted therapy or rehabilitation with increased focus on smoking and physical activity, may be contributing to international disparities in lung cancer survival. As the largest disparities were noted for regional stage disease, the opportunity for intervention may be greatest for this group. Finally, it is recommended that efforts should be made, and resources allocated, to improve the availability and comparability of stage data across countries. This will enable further research to understand the reasons behind international differences in stage-specific survival, which may fuel policy development and optimise prognosis.
Data availability statement
Data may be obtained from a third party and are not publicly available. Data contain sensitive and personal information and can be obtained from the cancer registries listed as the local leads in the manuscript.
Ethics statements
Patient consent for publication
Ethics approval
The study has been approved by the IARC Ethical Committee (meeting reference: EC 2016-04 and project reference: 16-36).
Acknowledgments
The authors would like to thank the ICBP management team of Cancer Research UK for managing the programme, the ICBP SURVMARK-2 Local Leads for advice to understand the data, for their contributions to the study protocol and interpretation of the results and the ICBP Clinical Committees for their advice. We are also grateful to the ICBP SURVMARK-2 Academic Reference Group for providing independent peer review and advice for the study protocol and analysis plan development. Finally, we are thankful to the ICBP Program Board for their oversight and direction.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
MA and MF-B contributed equally.
Collaborators ICBP SURVMARK-2 Local Leads: Alana Little (Cancer Institute NSW, Alexandria, New South Wales, Australia); Angela Eckstrand (Alberta Health Services, Edmonton, Alberta, Canada); Anna Gavin (Northern Ireland Cancer Registry, Queen’s University Belfast, Belfast, Northern Ireland, UK); Bin Zhang (Government of New Brunswick, Fredericton, New Brunswick, Canada); Bjørn Møller (Cancer Registry of Norway (Kreftregisteret), Oslo, Norway); Carol McClure (Government of Prince Edward Island, Charlottetown, Prince Edward Island, Canada); Rebecca Thomas (Welsh Cancer Intelligence and Surveillance Unit, Public Health Wales, Cardiff, Wales, UK); Connie Marano (Cancer Care Ontario, Toronto, Ontario, Canada); Christine Bertrand (Government of Quebec, Montreal, Quebec, Canada); Christopher Jackson (Cancer Society of New Zealand, Wellington, New Zealand); Claire Austin (Ministry of Health, New Zealand); David Gavin (The BC Cancer Research Centre, Vancouver, British Columbia, Canada); David S Morrison (Scottish Cancer Registry, University of Glasgow, Scotland); Deirdre Fitzpatrick (Northern Ireland Cancer Registry, Queen’s University Belfast, Belfast, Northern Ireland, UK); Donna Turner (Cancer Care Manitoba, Winnipeg, Manitoba, Canada); Dyfed Huws (Welsh Cancer Intelligence and Surveillance Unit, Public Health Wales, Cardiff, Wales, UK); Eshwar Kumar (Government of New Brunswick, Fredericton, New Brunswick, Canada); Gail Noonan (Cancer Care Manitoba, Winnipeg, Manitoba, Canada); Geoff Porter (Canadian Partnership Against Canada, Toronto, Ontario, Canada); Gerda Engholm (Surveillance and Pharmacoepidemiology, Danish Cancer Society Research Centre, Copenhagen, Denmark); Gina Lockwood (Canadian Partnership Against Canada, Toronto, Ontario, Canada); Grace Musto (Cancer Care Manitoba, Winnipeg, Manitoba, Canada); Grlica Bolesnikov (Government of New Brunswick, Fredericton, New Brunswick, Canada); Guillaume Ruel (Government of Quebec, Montreal, Quebec, Canada); Hanna Tervonen (Cancer Institute NSW, Alexandria, New South Wales, Australia); Hazem Abd Elkader (Cancer Society of New Zealand, Wellington, New Zealand); Heather Stuart-Panko (Saskatchewan Cancer Agency, Regina, Saskatchewan, Canada); Janet Warlow (Welsh Cancer Intelligence and Surveillance Unit, Public Health Wales, Cardiff, Wales, UK); Jason Poole (Public Health England, London, England, UK); John Spinelli (The BC Cancer Research Centre, Vancouver, British Columbia, Canada); Lorraine Shack (Alberta Health Services, Edmonton, Alberta, Canada); Louise Sandford (Cancer Society of New Zealand, Wellington, New Zealand); Marianne Brenn Jerm (Cancer Registry of Norway (Kreftregisteret), Oslo, Norway); Maureen MacIntyre (Nova Scotia Health Authority, Halifax, Nova Scotia, Canada); Nathalie St-Jacques (Nova Scotia Health Authority Cancer Care Program, Halifax, Nova Scotia, Canada); Oliver Bucher (Cancer Care Manitoba, Winnipeg, Manitoba, Canada); Paul Walsh (National Cancer Registry Ireland, Cork, Ireland); Prithwish De (Cancer Care Ontario, Toronto, Ontario, Canada); Raman Agnihotram (Research-Institude, McGill University Health Centre, Montreal, Quebec, Canada); Richard Trevithick (Department of Health, Perth, Western Australia, Australia); Rory Carle (Department of Health, Perth, Western Australia, Australia); Ryan Woods (Cancer Control Research, BC Cancer, Vancouver, British Columbia, Canada); Sally Vernon (Public Health England, London, England, UK); Serena Kozie (Saskatchewan Cancer Agency, Regina, Saskatchewan, Canada); Susan Ryan (Eastern Health, St. John’s, Newfoundland and Labrador, Canada); Sue Evans (Cancer Council Victoria, Melbourne, Victoria, Australia). ICBP SURVMARK-2 Academic Reference Group: Dianne O'Connell (Cancer Council New South Wales, Sydney, New South Wales, Australia); Michael Eden (Public Health England, London, UK); Mark Elwood (University of Auckland, New Zealand); Sabine Siesling (Netherlands Comprehensive Cancer Organisation (IKNL), Utrecht Amsterdam, The Netherlands; and University of Twente, Enschede, The Netherlands). ICBP Clinical Committee–Lung: Anur Guhan (The Ayr Hospital, Scotland, UK); Christian Finley (McMaster University, Canadian Association of Thoracic Surgeons, Canada); Brian McCaughan (University of Sydney, Australia); David Baldwin (University of Nottingham, UK); Erik Jakobsen (Odense University Hospital, Denmark); Ian Williamson (Aneurin Bevan University Hospital, Wales, UK); Jason Lester (Velindre Cancer Centre, Wales, UK); Jonathan McAleese (Belfast City Hospital, Northern Ireland, UK); Karl Kolbeck (Karolinska University Hospital, Sweden); Mick Peake (Public Health England, UK); Odd Terje Brustugun (Oslo University Hospital, Norway); Paul Dawkins (Middlemore Hospital, New Zealand); Wendy Anderson (Northern Health and Social Care Trust, Northern Ireland, UK).
Contributors MAr and MF-B analysed the data and drafted and revised the paper. MA, MR and AB wrote the statistical analysis plan, monitored data collection for the study and revised the draft paper. JF prepared the study protocol, monitored data collection and data harmonisation for the study, and revised the draft paper. OB, PD, GE, AG, SK, AL, BM, NSJ, HT, PW and RW prepared the study protocol, collected the data, contributed to the analytical plan and revised the draft of the paper. DLOC, DB, ME and SS commented on the analytical plan, study results and revised the draft of the paper. FB and IS designed the study, data collection tools, monitored data collection for the study, wrote the statistical analysis plan, and drafted and revised the paper. IS is the guarantor of the study.
Funding The project is funded by the International Cancer Benchmarking Partnership which is funded by the Canadian Partnership Against Cancer; Cancer Council Victoria; Cancer Institute New South Wales; Cancer Research UK; Danish Cancer Society; National Cancer Registry Ireland; The Cancer Society of New Zealand; NHS England; Norwegian Cancer Society; Public Health Agency Northern Ireland, on behalf of the Northern Ireland Cancer Registry; The Scottish Government; Western Australia Department of Health; and Wales Cancer Network.
Disclaimer Where authors are identified as personnel of the International Agency for Research on Cancer/WHO, the authors alone are responsible for the views expressed in this article and they do not necessarily represent the decisions, policy or views of the International Agency for Research on Cancer/WHO.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Linked Articles
- Airwaves