The Impact of COVID-19 on Lung Cancer Incidence in England
2023; Elsevier BV; Volume: 163; Issue: 6 Linguagem: Inglês
10.1016/j.chest.2023.01.008
ISSN1931-3543
AutoresSavannah Gysling, Helen Morgan, Onosi Sylvia Ifesemen, Douglas West, John Conibear, Neal Navani, Emma O’Dowd, David Baldwin, David J. Humes, Richard Hubbard,
Tópico(s)Economic and Financial Impacts of Cancer
ResumoBackgroundThe COVID-19 pandemic has caused significant disruption to health-care services and delivery worldwide. The impact of the pandemic and associated national lockdowns on lung cancer incidence in England have yet to be assessed.Research QuestionWhat was the impact of the first year of the COVID-19 pandemic on the incidence and presentation of lung cancer in England?Study Design and MethodsIn this retrospective observational study, incidence rates for lung cancer were calculated from The National Lung Cancer Audit Rapid Cancer Registration Datasets for 2019 and 2020, using midyear population estimates from the Office of National Statistics as the denominators. Rates were compared using Poisson regression according to time points related to national lockdowns in 2020.ResultsSixty-four thousand four hundred fifty-seven patients received a diagnosis of lung cancer across 2019 (n = 33,088) and 2020 (n = 31,369). During the first national lockdown, a 26% reduction in lung cancer incidence was observed compared with the equivalent calendar period of 2019 (adjusted incidence rate ratio [IRR], 0.74; 95% CI, 0.71-0.78). This included a 23% reduction in non-small cell lung cancer (adjusted IRR, 0.77; 95% CI, 0.74-0.81) and a 45% reduction in small cell lung cancer (adjusted IRR, 0.55; 95% CI, 0.46-0.65) incidence. Thereafter, incidence rates almost recovered to baseline, without overcompensation (adjusted IRR, 0.96; 95% CI, 0.94-0.98).InterpretationThe incidence rates of lung cancer in England fell significantly by 26% during the first national lockdown in 2020 and did not compensate later in the year. The COVID-19 pandemic has caused significant disruption to health-care services and delivery worldwide. The impact of the pandemic and associated national lockdowns on lung cancer incidence in England have yet to be assessed. What was the impact of the first year of the COVID-19 pandemic on the incidence and presentation of lung cancer in England? In this retrospective observational study, incidence rates for lung cancer were calculated from The National Lung Cancer Audit Rapid Cancer Registration Datasets for 2019 and 2020, using midyear population estimates from the Office of National Statistics as the denominators. Rates were compared using Poisson regression according to time points related to national lockdowns in 2020. Sixty-four thousand four hundred fifty-seven patients received a diagnosis of lung cancer across 2019 (n = 33,088) and 2020 (n = 31,369). During the first national lockdown, a 26% reduction in lung cancer incidence was observed compared with the equivalent calendar period of 2019 (adjusted incidence rate ratio [IRR], 0.74; 95% CI, 0.71-0.78). This included a 23% reduction in non-small cell lung cancer (adjusted IRR, 0.77; 95% CI, 0.74-0.81) and a 45% reduction in small cell lung cancer (adjusted IRR, 0.55; 95% CI, 0.46-0.65) incidence. Thereafter, incidence rates almost recovered to baseline, without overcompensation (adjusted IRR, 0.96; 95% CI, 0.94-0.98). The incidence rates of lung cancer in England fell significantly by 26% during the first national lockdown in 2020 and did not compensate later in the year. Take-home PointsStudy Question: What impact did the first year of the COVID-19 pandemic have on the incidence and presentation of lung cancer in England?Results: During the first national lockdown, a 26% (adjusted incidence rate ratio [IRR], 0.74; 95% CI, 0.71-0.78) reduction in lung cancer incidence was observed compared with the equivalent period in 2019, encompassing a 23% reduction (adjusted IRR, 0.77; 95% CI, 0.74-0.81) in non-small cell lung cancer and a 45% reduction (adjusted IRR, 0.55; 95% CI, 0.46-0.65) in small cell lung cancer incidence.Interpretation: The incidence rates of lung cancer in England fell significantly by 26% during the first national lockdown in 2020 and did not compensate later in the year. Study Question: What impact did the first year of the COVID-19 pandemic have on the incidence and presentation of lung cancer in England? Results: During the first national lockdown, a 26% (adjusted incidence rate ratio [IRR], 0.74; 95% CI, 0.71-0.78) reduction in lung cancer incidence was observed compared with the equivalent period in 2019, encompassing a 23% reduction (adjusted IRR, 0.77; 95% CI, 0.74-0.81) in non-small cell lung cancer and a 45% reduction (adjusted IRR, 0.55; 95% CI, 0.46-0.65) in small cell lung cancer incidence. Interpretation: The incidence rates of lung cancer in England fell significantly by 26% during the first national lockdown in 2020 and did not compensate later in the year. The COVID-19 pandemic has caused significant disruption to health-care services globally, with the full impact of this yet to be realized. More specifically, the effect on patients with lung cancer—the most common cause of cancer death in the United Kingdom1Cancer mortality for common cancers. Cancer Research UK. Accessed December 20, 2022. https://www.cancerresearchuk.org/health-professional/cancer-statistics/mortality/common-cancers-compared#heading-ZeroGoogle Scholar—has yet to be described fully. A report published by the United Kingdom Lung Cancer Coalition quotes an estimated 4.8% to 5.3% increase in 5-year mortality for patients with lung cancer as a consequence of the pandemic.2Maringe C. Spicer J. Morris M. et al.The impact of the COVID-19 pandemic on cancer deaths due to delays in diagnosis in England, UK: a national, population-based, modelling study.Lancet Oncol. 2020; 21: 1023-1034Abstract Full Text Full Text PDF PubMed Scopus (976) Google Scholar A key determinant of lung cancer outcomes is timely diagnosis, allowing the opportunity for rapid assessment and potential initiation of treatment for this aggressively progressive disease.3Tsai C.H. Kung P.T. Kuo W.Y. Tsai W.C. Effect of time interval from diagnosis to treatment for non-small cell lung cancer on survival: a national cohort study in Taiwan.BMJ Open. 2020; 10e034351Crossref Scopus (21) Google Scholar Similarly, a delay in the diagnosis of lung cancer can result in upstaging, fewer potential treatment options, and presumed worse survival outcome greater than that lost by the lead time.4Kanashiki M. Tomizawa T. Yamaguchi I. et al.Volume doubling time of lung cancers detected in a chest radiograph mass screening program: comparison with CT screening.Oncol Lett. 2012; 4: 513-516Crossref PubMed Scopus (29) Google Scholar,5Pujol J.L. Quantin X. Time to diagnosis of lung cancer: technical and pyschological factors that slow down diagnostic and treatment timelines.J Thorac Oncol. 2009; 4: 1192-1194Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar Currently, no national population-based lung cancer screening program exists to support early diagnosis in the United Kingdom, and lung cancer diagnosis commonly relies on the self-presentation of the patient or an incidental finding.6Adult screening programme lung cancer. UK National Screening Committee. Accessed December 20, 2022. https://view-health-screening-recommendations.service.gov.uk/lung-cancer/.Google Scholar The COVID-19 pandemic represented a unique barrier regarding patient access to lung cancer services and early diagnosis. Throughout several national lockdowns in England, the population was asked to remain at home and to self-isolate when new respiratory symptoms suggestive of viral infection emerged.7Stay at home: guidance for households with possible or confirmed coronavirus (COVID-19) infection. United Kingdom Health Security Agency. Accessed December 20, 2022. https://www.gov.uk/government/publications/covid-19-stay-at-home-guidance/stay-at-home-guidance-for-households-with-possible-coronavirus-covid-19-infection.Google Scholar,8Round T. L’Esperance V. Bayly J. et al.COVID-19 and the multidisciplinary care of patients with lung cancer: an evidence-based review and commentary.Brit J Cancer. 2021; 125: 629-640Crossref PubMed Scopus (17) Google Scholar Simultaneously, health-care resources were adapted, redeployed, and focussed on managing the COVID-19 pandemic. These factors are presumed to have affected patient access to health care, including diagnostic services. The extent of the impact of these measures and limitations on the access to health-care and diagnostic services is unclear. We sought to evaluate the impact of the COVID-19 pandemic on the incidence of lung cancer in England by analyzing the National Lung Cancer Audit 2019 and 2020 Rapid Cancer Registration Datasets. The National Cancer Registration and Analysis Service maintains the Cancer Analysis System, consisting of the Cancer Outcomes and Services Dataset, with linked data from the Hospital Episode Statistics dataset, Office for National Statistics, National Radiotherapy Dataset, and Systemic Anti-Cancer Dataset. This information is collated to produce annual Cancer Registration Datasets. For the years 2019 and 2020, these have been released in the form of Rapid Cancer Registration Datasets (RCRDs), with reduced sources (eg, lung cancers identified by death certificate registration alone), validation, and completeness compared with the usual finalized annual registrations for the sake of timely analysis and evaluation of the impact of COVID-19 on lung cancer in England. Data pertaining to all patients with a diagnosis of lung cancer in England were retrieved for audit within the framework of the nationally registered National Lung Cancer Audit. All adult patients with a diagnosis of incident primary lung cancer (as classified by code C34 of the World Health Organization’s International Classification of Diseases, Tenth Revision) in England between January 1, 2019, and December 31, 2020, were included. Patients with a diagnosis of sarcoma or mesothelioma were excluded. Patients were followed up from point of diagnosis to death or censoring. Midyear population estimates in England for 2019 and 2020 were derived from the Office for National Statistics9Population estimates for the UK, England and Wales, Scotland and Northern Ireland: mid-2020. UK Office for National Statistics. Accessed December 20, 2022. https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/bulletins/annualmidyearpopulationestimates/mid2020.Google Scholar for use as baseline population count. Any incidence rates per population are referring to these estimates, which also were used to calculate crude incidence per 100,000 person-years. Deprivation status was defined according to the Index of Multiple Deprivation 2019,10UK Ministry of HousingThe English Indices of Deprivation 2019 Research Report. UK Ministry of Housing, 2019Google Scholar scaled from least deprived (score of 1) to most deprived (score of 5). The Charlson comorbidity index (CCI)11Charlson M.E. Pompei P. Ales K.L. MacKenzie C.R. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation.J Chron Dis. 1987; 40: 373-383Abstract Full Text PDF PubMed Scopus (36499) Google Scholar was used to stratify the comorbidity burden of patients, with further categorization into subgroups zero (CCI 0), one (CCI 1), two (CC 2-3), and three (CCI 4 or more). Performance status was classified according to the World Health Organization Performance Status12Oken M.M. Creech R.H. Tormey D.C. et al.Toxicity and response criteria of the Eastern Cooperative Oncology Group.Am J Clin Oncol. 1982; 6: 649-655Crossref Scopus (8633) Google Scholar (scale of 0-4, best to worst). The eighth edition of the TNM Classification of Malignant Tumours was used in the staging of all patients’I diseases. Two ranges of TNM stages were used (IA-4B vs I-IV). The former demonstrates additional detail as to the patients’ staging, and the latter was used for comparison between time periods and includes patients for whom sublevel staging was not available. The year 2020 was divided into three periods (before lockdown, first lockdown, and after lockdown), based on instrumental political and governmental decisions.13Timeline of UK coronavirus lockdowns, March 2020 to March 2021. Institute for Government. Accessed December 20, 2022. https://www.instituteforgovernment.org.uk/charts/uk-government-coronavirus-lockdowns.Google Scholar The prelockdown period (T1) encompasses the time between January 1, 2020, and March 25, 2020. On March 26, 2020, the first and most stringent lockdown in England began. The period between this date and May 10, 2020 (lifting of the first lockdown restrictions), is defined as the first lockdown period (T2). Any time thereafter until the end of 2020 (December 31, 2020) is classified as the postlockdown period (T3). The first national lockdown was chosen as the focus of this study because the restrictions implemented during this lockdown period were the most stringent of those imposed throughout the year and public motivation for compliance was high. A descriptive summary of the baseline patient characteristics was conducted using mean ± SD for parametric data, median (interquartile range) for nonparametric data, and proportions for categorical variables. Because of the varying length of the COVID-19 periods within years, proportions and absolute number of patients were used for summative descriptions. Missing data were presented separately within summative tables. Continuous variables were evaluated using Student t test for parametric data and the Mann-Whitney U test for nonparametric data. Categorical patient variables (deprivation, comorbidity, performance status, and TNM stage) were assessed using the χ 2 or Fisher exact test, with comparison made across periods. Logistic regression models were used to evaluate overall change in the use of each referral pathway between years (using 2019 as the reference group), adjusted for sex and age. The 2-week wait pathway refers to the urgent referral pathway through which patients with suspected cancer are referred to secondary care in the United Kingdom. Results were summarized in the form of adjusted ORs as appropriate, with 95% CIs. Midyear population estimates in England were used for baseline population count in the calculation of crude incidence rates per 100,000 person-years. Monthly incidence rates were calculated using this baseline and were adjusted for person-time. Poisson regression was used to model the incidence rates in adjusted models, with results reported as adjusted incidence rate ratios (IRRs) per person-time. Results were stratified by the COVID-19 periods, with comparison of the three defined periods (T1, T2, and T3) in 2020 with the equivalent periods in 2019. Monthly incidence rates were analyzed, with each month in 2020 compared with the same period in 2019. For between-year (2019 vs 2020) comparisons, the 2019 cohort was used as a baseline. Subgroup analysis of patients with non-small cell lung cancer (NSCLC) and with small cell lung cancer (SCLC) was performed because of the clinical impact of these differential diagnoses. A P value of < .05 was considered statistically significant. All analyses were performed using Stata Statistical Software release 17 (StataCorp LLC). Sixty-four thousand four hundred fifty-seven patients (men, n = 33,349 [51.7%]) received a diagnosis of lung cancer across 2019 (n = 33,088) and 2020 (n = 31,369) was found. No difference was found in median age of those with a diagnosis of lung cancer between 2019 and 2020 (Table 1). In 2020, more data for smoking status were missing compared with 2019 (Table 1). No differences were observed in the age or sex of those with a diagnosis of either NSCLC or SCLC across the 2 years.Table 1Characteristics for All Patients With Lung Cancer in 2019 and 2020Variable20192020TotalAll lung cancer No. of patients33,08831,36964,457 Age, y73 (66-80)73 (66-80)73 (66-80) Male sex17,184 (51.9)16,165 (51.5)33,349 (51.7) Smoking statusSmoker6,785 (20.5)5,497 (17.5)12,282 (19.1)Former smoker11,551 (34.9)8,901 (28.4)20,452 (31.7)Nonsmoker351 (1.1)184 (0.6)535 (0.8)Never smoker1,552 (4.7)1,338 (4.3)2,890 (4.5)Declined20 (0.1)2 (0.0)22 (0.0)Missing12,829 (38.8)15,447 (49.2)28,276 (43.9)NSCLC No. of patients29,48028,33857,818 Age, y73 (66-80)74 (66-80)73 (66-80) Male sex15,469 (52.5)14,751 (52.1)30,220 (52.3)SCLC No. of patients2,9962,5555,551 Age, y70 (76-97)70 (76-97)70 (76-97) Male sex1,508 (50.3)1,245 (48.7)2,753 (49.6)Data are presented as No. (%) or median (interquartile range), unless otherwise indicated. NSCLC = non-small cell lung cancer; SCLC = small cell lung cancer. Open table in a new tab Data are presented as No. (%) or median (interquartile range), unless otherwise indicated. NSCLC = non-small cell lung cancer; SCLC = small cell lung cancer. A change in the use of referral pathways that lead to lung cancer diagnosis occurred in 2020. Adjusting for patient age and sex, emergency presentations (OR, 1.21; 95% CI, 1.18-1.26; P < .001) and other outpatient referrals (OR, 1.08; 95% CI, 1.04-1.13; P < .001) increased, whereas general practitioner referrals (OR, 0.89; 95% CI, 0.86-0.93; P < .001) and 2-week wait referrals (OR, 0.82; 95% CI, 0.79-0.85; P < .001) were reduced compared with 2019. No significant difference in inpatient elective referrals was observed (OR, 0.96; 95% CI, 0.85-1.09; P = .571) (Fig 1). The crude incidence rate of lung cancer diagnosis fell from 58.5 to 55.4 per 100,000 person-years in England in 2020 compared with 2019 (adjusted IRR, 0.95; 95% CI, 0.94-0.97). The distribution of lung cancers diagnosed in 2019 and 2020 across age groups is displayed in Figure 2. In the context of the pandemic and adjusting for age and sex, the incidence of lung cancer diagnosis during the prelockdown period was 4% higher (adjusted IRR, 1.04; 95% CI, 1.01-1.08) compared with the equivalent time period in 2019. During the first national lockdown in 2020, a 26% reduction in lung cancer incidence rates was observed compared with the equivalent period of the previous year (adjusted IRR, 0.74; 95% CI, 0.71-0.78). During the postlockdown period, incidence rates were only slightly lower than in the same period in 2019 (adjusted IRR, 0.96; 95% CI, 0.94-0.98) (Table 2). The monthly unadjusted incidence of lung cancer (per 100,000 person-years) and monthly adjusted (for age and sex) IRR for both years is displayed in Figure 3. The largest relative reduction in lung cancer incidence occurred during May 2020, representing a 32% reduction in incidence rates compared with the same month in 2019 (adjusted IRR, 0.68; 95% CI, 0.64-0.72).Table 2Incidence of Lung Cancer Across COVID-19 PeriodsVariableIR/100,000 Person-YAdjusted IRRaAdjusted for sex and age.95% CI20192020All lung cancer T158.5360.501.041.01-1.08 T259.3943.670.740.71-0.78 T358.5855.780.960.94-0.98NSCLC T152.3854.571.051.02-1.09 T252.5639.940.770.74-0.81 T352.1650.310.980.96-1.00SCLC T15.204.860.930.84-1.04 T25.483.020.550.46-0.65 T35.334.680.870.82-0.93NSCLC = non-small cell lung cancer; SCLC = small cell lung cancer; T1 = January 1-March 25 (prelockdown period in 2020); T2 = March 26-May 10 (first national lockdown period in 2020); T3 = May 11-December 31 (postlockdown period in 2020).a Adjusted for sex and age. Open table in a new tab Figure 3Bar graph showing the monthly IRs for all lung cancer cases in 2019 and 2020. The section highlighted in light pink represents the period of the first national lockdown. IR = incidence rate (unadjusted); Adjusted IRR = incidence rate ratio (adjusted for age and sex).View Large Image Figure ViewerDownload Hi-res image Download (PPT) NSCLC = non-small cell lung cancer; SCLC = small cell lung cancer; T1 = January 1-March 25 (prelockdown period in 2020); T2 = March 26-May 10 (first national lockdown period in 2020); T3 = May 11-December 31 (postlockdown period in 2020). The crude incidence rate of NSCLC fell from 52.37 per 100,000 person-years in 2019 to 50.11 per 100,000 person-years in 2020 (adjusted IRR, 0.97; 95% CI, 0.95-0.98). In patients with SCLC, the incidence rate fell from 5.32 per 100,000 person-years in 2019 to 4.52 per 100,000 person-years in 2020 (adjusted IRR, 0.85; 95% CI, 0.80-0.89). The incidence rates remained stable in the prelockdown period of 2020 compared with 2019 for both patients with NSCLC and those with SCLC (Table 2). During the first national lockdown, a 23% reduction in NSCLC incidence rates was observed compared with the equivalent period in 2019 (adjusted IRR, 0.77; 95% CI, 0.74-0.81). During the same period, a 45% reduction in SCLC incidence rates was observed (adjusted IRR, 0.55; 95% CI, 0.46-0.65) (Table 2). In the postlockdown period of 2020, incidence rates of NSCLC recovered (adjusted IRR, 0.98; 95% CI, 0.96-1.00) compared with the equivalent period in 2019, without compensation for the reduction observed during the first lockdown period. In the SCLC group, incidence remained substantially lower than baseline (adjusted IRR, 0.87; 95% CI, 0.82-0.93) (Table 2). The largest reduction in the monthly incidence rates occurred during April 2020 for patients with NSCLC (adjusted IRR, 0.68; 95% CI, 0.64-0.72) compared with April 2019 (e-Fig 1). For patients with SCLC, the greatest reduction occurred during April 2020 (adjusted IRR, 0.57; 95% CI, 0.46-0.70) (e-Fig 2). Patient demographic and performance characteristics across periods in 2019 and 2020 can be seen in Table 3 (all lung cancer) and in e-Table 1 (NSCLC) and e-Table 2 (SCLC). A difference in the performance status of patients with lung cancer was found during the first lockdown (P = .002) compared with the equivalent periods in 2019; however, this effect was negated on exclusion of patients with missing data, and the absolute differences between proportions across periods were small (Table 3). During the postlockdown period, a significant shift occurred in the performance status of lung cancer patients toward worsening (P < .001), which could not be accounted for by missing data alone, with fewer patients with a performance status of zero at presentation (Table 3).Table 3Characteristics of Patient With Lung Cancer Across COVID-19 Periods in 2019 and 2020VariableT1T2T3Deprivation (IMD 2019)201920202019202020192020 114.2 (1,089)14.1 (1,122)15.3 (644)13.8 (430)14.1 (3,000)14.8 (3,002) 218.6 (1,424)19.3 (1,534)18.7 (788)18.7 (581)18.4 (3,910)18.6 (3,778) 319.8 (1,516)19.9 (1,586)18.7 (788)20.2 (629)20.1 (4,270)19.7 (3,999) 421.4 (1,641)21.3 (1,692)22.0 (928)21.3 (663)21.7 (4,595)21.5 (4,371) 526.0 (1,994)25.5 (2,028)25.2 (1,063)26.0 (809)25.6 (5,438)25.4 (5,145) Total100 (7,664)100 (7,962)100 (4,211)100 (3,112)100 (21,213)100 (20,295)Comorbidity201920202019202020192020aStatistically significant for P < .01, 2020 vs 2019. 065.3 (5,006)64.8 (5,161)65.5 (2,756)66.4 (2065)65.2 (13,821)65.7 (13,342) 113.8 (1,054)14.7 (1,167)13.5 (567)14.3 (444)13.9 (2,957)14.6 (2,970) 214.8 (1,132)14.6 (1,165)15.3 (643)13.2 (411)15.0 (3,188)13.9 (2,819) 36.2 (472)5.9 (469)5.8 (245)6.2 (192)6.0 (1,247)5.7 (1,164) Total100 (7,664)100 (7,962)100 (4,211)100 (3,112)100 (21213)100 (20,295)Performance status20192020bStatistically significant for P < .05, 2020 vs 2019.20192020aStatistically significant for P < .01, 2020 vs 2019.20192020cStatistically significant for P < .001, 2020 vs 2019. 019.6 (1,504)20.4 (1,627)20.8 (877)18.9 (587)20.3 (4,299)16.7 (3,391) 132.0 (2,545)30.4 (2,419)32.2 (1,354)30.3 (943)31.3 (6,635)28.2 (5,718) 216.5 (1,264)16.6 (1,322)15.8 (666)16.2 (504)16.7 (3,536)16.9 (3,433) 316.5 (1,261)16.6 (1,323)16.6 (700)17.8 (553)16.3 (3,448)17.8 (3,613) 45.7 (436)5.0 (399)5.3 (225)5.1 (158)4.8 (1,023)5.8 (1,181) Missing9.7 (745)11.0 (872)9.2 (389)11.8 (367)10.7 (2,272)14.6 (2,959) Total100 (7,664)100 (7,962)100 (4,211)100 (3,112)100 (21,213)100 (20,295)TNM stage20192020aStatistically significant for P < .01, 2020 vs 2019.2019202020192020cStatistically significant for P < .001, 2020 vs 2019. I19.0 (1,458)21.7 (1,731)19.5 (822)19.6 (611)20.2 (4,285)19.1 (3,882) II8.2 (626)8.2 (654)8.1 (340)8.1 (252)8.1 (1,709)6.2 (1,266) III21.7 (1,663)20.6 (1,642)21.4 (903)19.1 (593)21.3 (4,517)18.5 (3,763) IV43.4 (3,324)41.5 (3,303)44.0 (1,853)45.4 (1,413)42.9 (9,106)45.2 (9,163) Missing7.7 (593)7.9 (632)7.0 (293)7.8 (243)7.5 (1,596)10.9 (2,221) Total100 (7,664)100 (7,962)100 (4,211)100 (3,112)100 (21,213)100 (20,295)Data are presented as % (No.). IMD = Index of Multiple Deprivation; T1 = January 1-March 25 (prelockdown period in 2020); T2 = March 26-May 10 (first national lockdown period in 2020); T3 = May 11-December 31 (postlockdown period in 2020).a Statistically significant for P < .01, 2020 vs 2019.b Statistically significant for P < .05, 2020 vs 2019.c Statistically significant for P < .001, 2020 vs 2019. Open table in a new tab Data are presented as % (No.). IMD = Index of Multiple Deprivation; T1 = January 1-March 25 (prelockdown period in 2020); T2 = March 26-May 10 (first national lockdown period in 2020); T3 = May 11-December 31 (postlockdown period in 2020). During the first national lockdown, no change in TNM stage distribution was observed compared with that in 2019 (P = .109) (Table 3). After the first lockdown was lifted, a stage shift was observed (P < .001) compared with the equivalent period in 2019, with reduced stage I (mean difference [MD], –1.1%), stage II (MD, –1.8%), and stage III (MD, –2.8%) presentations and increased stage IV (MD, 2.2%) presentations in the context of increased missing data in 2020 (MD, 3.4%) (Table 3). Further description of the TNM stage distribution across patients with NSCLC and SCLC can be found in e-Tables 3 and 4. Our analysis of the 2019 and 2020 RCRDs shows a 26% reduction in lung cancer incidence during the first national lockdown in 2020, with a 23% decrease in NSCLC and 45% decrease in SCLC incidence rates during this period. Given the relatively stable incidence rate of lung cancer in the United Kingdom in the preceding decade,14Lung cancer incidence statistics. Cancer Research UK. Accessed December 20, 2022. https://www.cancerresearchuk.org/health-professional/cancer-statistics/statistics-by-cancer-type/lung-cancer/incidence#heading-Two.Google Scholar this represents a novel and unprecedented change. Evidence was found of stage migration and worsening performance status of patients with lung cancer after the first lockdown period. Although ambulatory services for patients with lung cancer remained largely unchanged and available throughout the pandemic, multiple factors may have impeded access to these, particularly for patients with lung cancer who tend to seek treatment at an older age with multiple comorbidities, such as fear of contracting COVID-19, lockdown regulations restricting movement, and limitations in transport availability. Decreased routine care for these comorbidities and therefore increased overall morbidity, as reflected in the decrease in performance status observed, may have contributed to the reduced lung cancer incidence. Given the lack of compensation of the incidence rates after lockdown, as well as the stage shift observed, these so-called missing cases also may be attributable in part to patients who died without receiving a diagnosis. This is supported by our findings that the fall in incidence was greater for SCLC, a more aggressive cancer, in which patients have less time to seek treatment before they die. An Italian multicenter study15Cantini L, Mentrasti G, Siena S. Impact of COVID19 outbreak on lung cancer diagnosis and continuum of care: data from an Italian multicenter study. Paper presented at: IASLC 2021 World Conference on Lung Cancer Hosted by the International Association for the Study of Lung Cancer; August 9, 2021.Google Scholar assessed the incidence of lung cancer in 2020 compared with the previous year. They reported a 6.9% reduction in lung cancer diagnosis in 2020 compared with 2019, with the maximum reduction occurring during May 2020. In a two-center study from Spain, Reyes et al16Reyes R. López-Castro R. Auclin E. et al.MA03.08 Impact of COVID-19 pandemic in the diagnosis and prognosis of lung cancer.J Thorac Oncol. 2021; 16 (S141-S141)Abstract Full Text Full Text PDF Google Scholar found a 38% reduction in lung cancer cases diagnosed during the defined COVID-19 period compared with baseline. Kasymjanova et al17Kasymjanova G. Anwar A. Cohen V. et al.The impact of COVID-19 on the diagnosis and treatment of lung cancer at a Canadian academic center: a retrospective chart review.Curr Oncol. 2021; 28: 4247-4255Crossref PubMed Scopus (17) Google Scholar report a 21% reduction in lung cancer diagnosis in 2020 compared with 2019 at a single Canadian institution. These collective reports of reduced lung cancer diagnoses resulting from the pandemic are limited in part by the variation in the periods analyzed, the inability to adjust for potential confounding by sex and age, and the small sample sizes. Furthermore, the management and adaptation of health-care services during the pandemic has varied widely across regions and nations, as has the timeline of COVID-19 peaks and national lockdowns, which undoubtably have impacted lung cancer diagnosis differentially. To our knowledge, our study represents the first national description of the impact of lockdowns on the incidence of lung cancer. Importantly, the overall lung cancer diagnostic services were able to recover after the first national lockdown and throughout the subsequent second lockdown in England. The latter were less stringent in their restrictions and allowed for tiered, regional variation, dependant on COVID-19 case numbers. It is possible that these factors, as well as growing experience with the novel virus, allowed the diagnostic services to adapt and more patients to seek treatment. As the pandemic continues, it is important that these services be preserved, with both patients and health-care services continuing to prioritize the early diagnosis and screening for lung cancer. Few reports exist on the impact of the pandemic on the TNM stage of patients with lung cancer at diagnosis. Our study found a significant shift toward patients seeking treatment with stage IV lung cancer after the first national lockdown. Cantini et al15Cantini L, Mentrasti G, Siena S. Impact of COVID19 outbreak on lung cancer diagnosis and continuum of care: data from an Italian multicenter study. Paper presented at: IASLC 2021 World Conference on Lung Cancer Hosted by the International Association for the Study of Lung Cancer; August 9, 2021.Google Scholar observed a similar increase in the proportion of patients with lung cancer seeking treatment with stage IV lung cancer in 2020 compared with 2019 (2019, 67% vs 2020, 72%; P < .01). Similarly, Carroll et al18Carroll H. Doyle M. Eaton D. et al.FP06.02 the impact of the COVID-19 pandemic on new diagnoses of lung cancer: a 3-year review of an Irish cancer centre.J Thorac Oncol. 2021; 16 (S954-S954)Abstract Full Text Full Text PDF Google Scholar reported a shift toward stage IV disease in patients with NSCLC during the pandemic (34.4% vs 46.3%; P = .01), whereas Reyes et al16Reyes R. López-Castro R. Auclin E. et al.MA03.08 Impact of COVID-19 pandemic in the diagnosis and prognosis of lung cancer.J Thorac Oncol. 2021; 16 (S141-S141)Abstract Full Text Full Text PDF Google Scholar found more advanced NSCLC disease at presentation during the COVID-19 period. It is known that lung cancer may progress rapidly, with a median volume doubling time of approximately 104 days4Kanashiki M. Tomizawa T. Yamaguchi I. et al.Volume doubling time of lung cancers detected in a chest radiograph mass screening program: comparison with CT screening.Oncol Lett. 2012; 4: 513-516Crossref PubMed Scopus (29) Google Scholar (average of 70 days in SCLC),19Sharial M.S.N.M. Teo M. Doherty M. McDermott R.S. Walshe J.M. Modern imaging technique assessment of small cell lung cancer doubling time.J Clin Oncol. 2012; 30 (e17561-e17561)Google Scholar depending on cell type, disease stage, and patient factors. Any delay in diagnosis therefore is assumed to have critical effects on tumor upstaging, as well as in the determination of subsequent treatment options and outcomes.2Maringe C. Spicer J. Morris M. et al.The impact of the COVID-19 pandemic on cancer deaths due to delays in diagnosis in England, UK: a national, population-based, modelling study.Lancet Oncol. 2020; 21: 1023-1034Abstract Full Text Full Text PDF PubMed Scopus (976) Google Scholar Lockdowns may have led to reduced physical activity, particularly in the elderly, and resultant physical and mental deconditioning.20Public Health EnglandWider Impacts of COVID-19 on Physical Activity, Deconditioning and Falls in Older Adults. Public Health England, 2021Google Scholar This could explain the shift toward worse performance status at presentation observed in our analysis. Carroll et al18Carroll H. Doyle M. Eaton D. et al.FP06.02 the impact of the COVID-19 pandemic on new diagnoses of lung cancer: a 3-year review of an Irish cancer centre.J Thorac Oncol. 2021; 16 (S954-S954)Abstract Full Text Full Text PDF Google Scholar did not find any significant shift in the performance status of 491 patients with lung cancer seeking treatment over a 12-month period (December 2019-November 2020), although the potential shift may have been too small to detect in this population. The performance status is vital in determining both available management options for patients with lung cancer, as well as outcome with treatment. This may present an area for targeted intervention to improve outcomes for patients with lung cancer. The RCRDs provide an overview of patients who receive a diagnosis of lung cancer in England, capturing changes affecting these patients at a national level. The strength of this work lies not only in the population size encompassed, but also in the range of patient and health-care factors yielded by the combined use of several linked datasets. Through the release of these newly developed RCRDs, we were able to perform a timely analysis of the evolving impact of the COVID-19 pandemic, which may inform the ongoing response to the pandemic. The newly developed RCRDs allow for timely analysis of the changes in lung cancer diagnosis brought on by the pandemic, albeit at the cost of reduced validation and follow-up time. For example, the large proportion of missing values for smoking status precluded further analysis of this factor beyond a yearly analysis. Furthermore, lung cancers identified by death certificate registration alone were not included in the RCRDs and may account selectively for some of the missing cases. Because these datasets are a preliminary component of the eventual gold standard registration data, our results may not match subsequent National Statistics publications. They also represent a retrospective viewpoint on a rapidly evolving situation with a limited follow-up period, potentially leading to underestimation of the true impact of the pandemic on patients with lung cancer. Full details regarding the quality of these datasets are available at www.ncin.org.uk/collecting_and_using_data/rcrd. We acknowledge that dividing the years into COVID-19 periods results in an unequal distribution of the absolute number of patients per period and relatively small group numbers in the SCLC cohort. This may be reflected in less precise crude incidence rates and wider CIs surrounding the adjusted IRRs pertaining to this group. The interpretation of the comparative patient characteristics therefore was focussed on proportions rather than absolute numbers. The incidence of lung cancer has been affected significantly by the COVID-19 pandemic and associated societal and health-care changes, with those with SCLC most disadvantaged. The pandemic has illustrated what happens when access to prompt diagnosis is severely limited. As the pandemic continues, it is vital that awareness campaigns are enhanced and linked to prompt access to clinical triage, early diagnosis, and appropriate treatment to mitigate excess mortality from a cancer that causes almost one-fifth of cancer deaths. N. N. is supported by a Medical Research Council Clinical Academic Research Partnership [Grant MR/T02481X/1]. Work relating to the National Lung Cancer Audit was undertaken partly at the University College London Hospitals/University College London and received a proportion of funding from the Department of Health’s National Institute for Health Research Biomedical Research Centre’s funding scheme.
Referência(s)