Isolating the effect of confounding from the observed survival benefit of screening participants - a methodological approach illustrated by data from the German mammography screening programme.

BACKGROUND: Mammography screening programmes (MSP) aim to reduce breast cancer mortality by shifting diagnoses to earlier stages. However, it is difficult to evaluate the effectiveness of current MSP because analyses can only rely on observational data, comparing women who participate in screening with women who do not. These comparisons are subject to several biases: one of the most important is self-selection into the MSP, which introduces confounding and is difficult to control for. Here, we propose an approach to quantify confounding based on breast cancer survival analyses using readily available routine data sources. METHODS: Using data from the Cancer Registry of North Rhine-Westphalia, Germany, we estimate the relative contribution of confounding to the observed survival benefit of participants of the German MSP. This is accomplished by comparing non-participants, participants with screen-detected and participants with interval breast cancers for the endpoints "death from breast cancer" and "death from all causes other than breast cancer" - the latter being assumed to be unrelated to any MSP effect. By using different contrasts, we eliminate the effects of stage shift, lead and length time bias. The association of breast cancer detection mode with survival is analysed using Cox models in 68,230 women, aged 50-69 years, with breast cancer diagnosed in 2006-2014 and followed up until 2018. RESULTS: The hazard of dying from breast cancer was lower in participants with screen-detected cancer than in non-participants (HR = 0.21, 95% CI: 0.20-0.22), but biased by lead and length time bias, and confounding. When comparing participants with interval cancers and non-participants, the survival advantage was considerably smaller (HR = 0.62, 95% CI: 0.58-0.66), due to the elimination of stage shift and lead time bias. Finally, considering only mortality from causes other than breast cancer in the latter comparison, length time bias was minimised, but a survival advantage was still present (HR = 0.63, 95% CI: 0.56-0.70), which we attribute to confounding. CONCLUSIONS: This study shows that, in addition to stage shift, lead and length time bias, confounding is an essential component when comparing the survival of MSP participants and non-participants. We further show that the confounding effect can be quantified without explicit knowledge of potential confounders by using a negative control outcome.

Mortality from prostate cancer in the years 2007-2021 in North Rhine-Westphalia, Germany.

BACKGROUND: The crude mortality rate and the lifetime mortality risk from prostate cancer in Germany are above international average. However age-standardised mortality and years of life lost per capita from prostate cancer are declining. This study analyses the mortality-related measures for the federal state of North Rhine-Westphalia (NRW) in Germany. METHODS: Based on the cause of death statistics and data from the NRW State Cancer Registry on 45,300 deaths in the years 2007-2021, mortality rates, the lifetime mortality risk from prostate cancer, median age at death and years of life lost are presented. Additionally, the 15 most frequent causes of death of 95,013 patients diagnosed with prostate cancer are reported. RESULTS: With a stable lifetime mortality risk from prostate cancer, age-standardised mortality and years of life lost per capita are decreasing while crude mortality and median age at death are increasing in NRW. Less than half of the patients die from their prostate cancer. Cancers of the urinary bladder and other urinary organs also occur more frequently as a cause of death than it would be expected based on the age-specific risk in the total population. CONCLUSIONS: More people in North Rhine-Westphalia are dying of prostate cancer over time due to demographic ageing alone. At the same time, the age-specific mortality risk has not increased and when patients die of prostate cancer, it is at an increasingly older age. However, there is a statistical association with deaths from cancers of the lower urinary tract in patients diagnosed with prostate cancer, which demands further evaluation.

Time trends in survival and causes of death in multiple myeloma: a population-based study from Germany.

BACKGROUND: Steady evolution of therapies has improved prognosis of patients with multiple myeloma (MM) over the past two decades. Yet, knowledge about survival trends and causes of death in MM might play a crucial role in long-term management of this patient collective. Here, we investigate time trends in myeloma-specific survival at the population level over two decades and analyse causes of death in times of prolonged survival. METHODS: Age-standardised and age group-specific relative survival (RS) of MM patients aged < 80 years at diagnosis was estimated for consecutive time periods from 2000-2019 using data from the Cancer Registry of North Rhine-Westphalia in Germany. Conditional RS was estimated for patients who already survived one to five years post diagnosis. Causes of death in MM patients were analysed and compared to the general population using standardised mortality ratios (SMR). RESULTS: Three thousand three hundred thirty-six MM cases were included in the time trend analysis. Over two decades, age-standardised 5-year RS increased from 37 to 62%. Age-specific survival improved from 41% in period 2000-2004 to 69% in period 2015-2019 in the age group 15-69 years, and from 23 to 47% in the age group 70-79 years. Conditional 5-year RS of patients who survived five years after diagnosis slightly improved as compared to unconditional 5-year RS at diagnosis. MM patients are two times more likely to die from non-myeloma malignancies (SMR = 1.97, 95% CI 1.81-2.15) and from cardiovascular diseases (SMR = 2.01, 95% CI 1.86-2.18) than the general population. CONCLUSIONS: Prognosis of patients with MM has markedly improved since the year 2000 due to therapeutic advances. Nevertheless, late mortality remains a major concern. As survival improves, second primary malignancies and cardiovascular events deserve increased attention.

Differences in site-specific incidence and relative survival of cutaneous and mucocutaneous genital squamous cell carcinoma in Germany, 2007-2015.

Direct comparisons of the incidence and survival of cutaneous vs mucocutaneous genital squamous cell carcinomas (SCCs) are lacking even though they may bring important insights. We aimed to compare incidence rates and survival of cutaneous and mucocutaneous genital SCCs head-to-head, using the same source population, cancer registry methodology and statistical methods in a population of predominantly white Caucasian descent. Using data (2007-2015) from the population-based cancer registry of North Rhine-Westphalia, (population of 18 million people), we estimated age-specific and age-standardized (old European standard) incidence rates and age-standardized relative 5-year survival of SCC with the period approach for the period 2012 to 2015. Overall, 83 650 SCC cases were registered. The age-standardized incidence rates (per 100 000 person-years) of cutaneous SCCs were 36.5 (SE 0.17) and 17.0 (SE 0.11) among men and women, respectively, with corresponding rates for mucocutaneous genital skin, 1.3 (SE 0.03) and 4.5 (SE 0.06) for men and women, respectively. In all age groups, incidence rates of mucocutaneous genital SCCs were higher in women than men. Men had higher cutaneous SCC incidence at all nongenital subsites than women, with the exception of the lower extremities. Five-year relative survival was considerably lower for mucocutaneous genital SCCs (men: 71%, women: 75%), especially of the scrotal skin (67%) and labia majora (62%) than for SCC of nongenital skin (men: 93%, women: 97%). Given their relatively high incidence together with a lower survival probability, future studies are warranted to establish therapies for advanced mucocutaneous genital SCC, such as immune checkpoint inhibition.

Incidence of advanced-stage breast cancer in regular participants of a mammography screening program: a prospective register-based study.

BACKGROUND: The European Guidelines for breast cancer screening suggest that the impact of population-based mammography screening programmes (MSP) may be assessed using the relative reduction in the incidence of advanced breast cancer (ABC, that is, stage UICC II and higher) as a surrogate indicator of screening effectiveness. METHODS: This prospective, population register-based study contained individual data of 1,200,246 women (aged 50-69 years) who attended the initial prevalence screening between 2005 and 2009. Of them, 498,029 women returned for the regular (i.e., within 24 months) first subsequent, and 208,561 for the regular second subsequent incidence screenings. The incidence rate of ABC was calculated for the 24-months period following, but not including, the initial screening by incorporating all interval ABCs and all ABCs detected at the regular first incidence screening; the ABC rate for the second 24-months period was determined in the same way, including ABCs detected in the interval after the first and, respectively, at the second incidence screening. The relative reduction in the ABC incidence was derived by comparing the age-standardized rates in these two periods with an age-standardized reference incidence rate, observed in the target population before the MSP implementation. The strengths and weaknesses of this particular study design were contrasted with a recently published checklist of main methodological problems affecting studies of the effect of MSP on ABC incidence. RESULTS: The age-standardized ABC incidence rate was 291.6 per 100,000 women for the 24-months period subsequent to the initial screening, and 275.0/100,000 for the 24-months period following the first subsequent screening. Compared to the 2-year incidence of 349.4/100,000 before the start of the MSP, this amounted to a relative reduction of 16.5 and 21.3%, respectively, in the incidence of ABC among regular MSP participants. CONCLUSIONS: The design employed in this study avoids some of the substantial methodological limitations that compromised previous observational studies. Nevertheless, specific limitations prevail that demand a cautious interpretation of the results. Therefore, the study findings, indicating a reduction in ABC for regular MSP participants, need to be followed with respect to potential impacts on breast cancer mortality rates.

Incidence and Relative Survival of Patients with Merkel Cell Carcinoma in North Rhine-Westphalia, Germany, 2008-2021.

BACKGROUND: To date, only a few population-representative studies have been carried out on the rare Merkel cell carcinoma (MCC). We provide incidence and survival estimates of MCC, including the conditional relative survival. METHODS: We analyzed data from the cancer registry of North Rhine-Westphalia, Germany, 2008-2021, covering a population of 18 million. We included all newly diagnosed MCCs and calculated age-standardized (old European Standard population) incidence rates and unconditional and conditional relative survival. RESULTS: Our analysis included 2164 MCC patients. The age-standardized incidence of MCC was 5.2 (men) and 3.8 (women) per million person-years. The 5-year relative survival was 58.8% (men) and 70.7% (women). Survival was lower among men than women in all age-sex groups and was highest for MCC of the upper extremity in both men (68.2%) and women (79.3%). The sex difference in survival is particularly due to the better survival of women with MCC of the head and neck. In terms of survival, the first two years are particularly critical. CONCLUSIONS: Our data validate the worse survival among men and highlights a more favorable prognosis for MCCs located on the limbs. The first two years after diagnosis of MCC are the years with the highest excess mortality.

The Epidemiology of Colorectal Cancer in Younger and Older Patients.

BACKGROUND: Recent studies have shown that the incidence of colorectal cancer among younger persons is rising. We investigated incidence trends and survival in the German federal state of North Rhine-Westphalia. METHODS: Cancer registry data from the period 2008-2019 were classified into two age groups (15-54 and 55-99). In each age group, the standardized incidence, average annual percent change (AAPC), and relative 5-year survival (RS) were calculated and stratified according to the site, histology, size, and grade of the colorectal tumor. RESULTS: 167 919 cases were included, with adenocarcinoma accounting for 86.4%. In 2019, the age-standardized incidence per 100 000 person-years was 13.8 and 10.3 among men and women (respectively) in the younger age group, compared with 197.9 and 126.3 in the older age group. Over the study period from 2008 to 2019, the incidence declined among older men and women (AAPC -2.6% and -2.9%) but remained nearly constant among younger men and women (-0.5% and -0.4%). The incidence of neuroendocrine, T1, and G2 tumors rose in both age groups (AAPC range: 2.3%-12.2%; 2.2%-8.3%, 6.3%-8.8%). Younger patients have a better RS, with the largest difference between age groups being found for neuroendocrine tumors (88% and 83% in younger men and women, 65% and 61% in older men and women). CONCLUSION: The incidence of colorectal tumors has remained constant in persons under age 55 and declined in persons aged 55 and older. Nonetheless, the incidence of neuroendocrine tumors and of small and well-differentiated tumors has risen in both age groups. The trends among younger persons and the rise in neuroendocrine tumors merit further study.

Incidence and relative survival of pancreatic adenocarcinoma and pancreatic neuroendocrine neoplasms in Germany, 2009-2018. An in-depth analysis of two population-based cancer registries.

BACKGROUND: Pancreatic neuroendocrine neoplasms are categorized as neuroendocrine tumors and neuroendocrine carcinomas. Until now, cancer registry reporting of pancreatic cancers does not include a stratification by these two subgroups. We studied the incidence and survival of pancreatic cancer with a special focus on pancreatic neuroendocrine neoplasms. METHODS: We analyzed data from the population-based cancer registries of North Rhine-Westphalia (NRW) and Saarland (SL), Germany, of the years 2009-2018. We included primary malignant pancreatic tumors and report morphology-specific age-standardized (World Standard population) incidence rates for ages 0-79 years and age-standardized relative survival (period approach, ICSS standard). All analyses were restricted to non-death certificate only cases. RESULTS: We analyzed 23,037 patients with a newly diagnosed primary pancreatic cancer. Among morphologically specified cancers, adenocarcinoma (92 %) and neuroendocrine neoplasms (7 %) were the most common morphologies. The age-standardized incidence rates of adenocarcinoma, neuroendocrine tumors and neuroendocrine carcinomas were 4.0-5.5 (in NRW and SL), 0.1-0.3, and 0.1-0.3 per 100,000 person-years, respectively. Neuroendocrine tumors had the highest age-standardized 5-year relative survival with 75.5 % (standard error, SE 2.3) in NRW and 90.6 % (SE 10.2) in SL followed by neuroendocrine carcinomas (NRW: 30.0 %, SE 3.1; SL: 32.3 %, SE 8.7) and adenocarcinomas (NRW: 11.3 %, SE 0.4; SL: 10.2 %, SE 1.5). DISCUSSION: The distinction between neuroendocrine tumors and neuroendocrine carcinomas by the WHO divides neuroendocrine neoplasms into two prognostically clearly distinct subgroups that should be separately analyzed in terms of survival. The first year after diagnosis of pancreatic cancer is the most critical year in terms of survival.

Survival of patients with chronic lymphocytic leukemia before and after the introduction of chemoimmunotherapy in Germany.

Chronic lymphocytic leukemia (CLL) is the most common leukemia of adults in western countries. Therapy is indicated in symptomatic and advanced stages and has changed fundamentally since 2010 when rituximab, an anti-CD20 antibody, has been approved for treatment of CLL. Until then therapy had been based on chemotherapy drugs. This study investigates whether survival in CLL patients improved at the population level after the introduction of combined chemoimmunotherapy. Data from the cancer registry North-Rhine Westphalia was used to calculate relative survival (RS) by applying period analyses. Age-standardized 5-year RS increased from 79% in 1998-2002 (75% in 2003-2007) to 81% in the calendar period 2008-2012 and 88% in 2013-2016 for men and continuously from 71% in 1998-2002 to 92% in 2013-2016 for women. In CLL patients aged 15-69 years 5-year RS increased from 83% to 90% for men and from 82% to 94% for women after adding an anti-CD20-antibody to chemotherapy while in the older age group of 70-79-year-old CLL patients an increase by 20 percentage points was observed. These findings show marked improvements in the survival of CLL patients at the population level subsequently to the approval of anti-CD 20 antibodies like rituximab, ofatumumab or obinutuzumab for CLL treatment.

Trends in advanced breast cancer incidence rates after implementation of a mammography screening program in a German population.

BACKGROUND: Mammography screening programs (MSPs) aim to detect early-stage breast cancers in order to decrease the incidence of advanced-stage breast cancers and to reduce breast cancer mortality. We analyzed the time trends of advanced-stage breast cancer incidence rates in the target population before and after implementation of the MSP in a region of northwestern Germany. METHODS: The MSP in the Münster district started in October 2005. A total of 13,874 women with an incident invasive breast cancer (BC) was identified by the population-based epidemiological cancer registry between 2000 and 2013 in the target group 50-69 years. Multiple imputation methods were used to replace missing data on tumor stages (10.4%). The incidence rates for early-stage (UICC I) and advanced-stage (UICC II+) BC were determined, and Poisson regression analyses were performed to assess trends over time. RESULTS: The incidence rates for UICC I breast cancers increased during the step-up introduction of the MSP and remained elevated thereafter. By contrast, after increasing from 2006 to 2008, the incidence rates of UICC II+ breast cancers decreased to levels below the pre-screening period. Significantly decreasing UICC II+ incidence rates were limited to the age group 55-69 years and reached levels that were significantly lower than incidence rates in the pre-screening period. DISCUSSION: The incidence rates of advanced-stage breast cancers decreased in the age groups from 55 years to the upper age limit for screening eligibility, but not in the adjacent age groups. The findings are consistent with MSP lead time effects and seem to indicate that the MSP lowers advanced-stage breast cancer rates in the target population.