Survival analysis for lung cancer patients: A comparison of Cox regression and machine learning models.

INTRODUCTION: Survival analysis based on cancer registry data is of paramount importance for monitoring the effectiveness of health care. As new methods arise, the compendium of statistical tools applicable to cancer registry data grows. In recent years, machine learning approaches for survival analysis were developed. The aim of this study is to compare the model performance of the well established Cox regression and novel machine learning approaches on a previously unused dataset. MATERIAL AND METHODS: The study is based on lung cancer data from the Schleswig-Holstein Cancer Registry. Four survival analysis models are compared: Cox Proportional Hazard Regression (CoxPH) as the most commonly used statistical model, as well as Random Survival Forests (RSF) and two neural network architectures based on the DeepSurv and TabNet approaches. The models are evaluated using the concordance index (C-I), the Brier score and the AUC-ROC score. In addition, to gain more insight in the decision process of the models, we identified the features that have an higher impact on patient survival using permutation feature importance scores and SHAP values. RESULTS: Using a dataset including the cancer stage established by the Union for International Cancer Control (UICC), the best performing model is the CoxPH (C-I: 0.698±0.005), while using a dataset which includes the tumor size, lymph node and metastasis status (TNM) leads to the RSF as best performing model (C-I: 0.703±0.004). The explainability metrics show that the models rely on the combined UICC stage and the metastasis status in the first place, which corresponds to other studies. DISCUSSION: The studied methods are highly relevant for epidemiological researchers to create more accurate survival models, which can help physicians make informed decisions about appropriate therapies and management of patients with lung cancer, ultimately improving survival and quality of life.

Initial Evidence of a Possible Effect of HPV Vaccination on Cancer Incidence in Germany.

BACKGROUND: Infection with human papillomavirus (HPV) is one of the more common sexually transmitted diseases in Germany. Vaccination against HPV was introduced in Germany in 2007. In this study, we sought to detect a population-based decline in the incidence of cervical cancer in women under age 30 who were eligible for vaccination in the first 11 years after its introduction. METHODS: Data on new diagnoses of HPV-associated cervical cancer from 2004 to 2018 were obtained from the cancer registries of the German federal states (Bundesländer) through the German Center for Cancer Registry Data (ZfKD). Trends in the incidence of invasive and in situ cervical cancer were determined with log-linear joinpoint regression and age-period-cohort models. RESULTS: The incidence of cervical cancer, which had been rising in the previous decades, has been falling since 2010, with a marked decline among women in all age groups eligible for vaccination (e.g., from 70.0 to 41.8 cases per 100 000 persons per year from 2010 to 2018 in women aged 24 to 26). Women born in 1992 were the first to become eligible for vaccination and have a 24% lower incidence than the reference cohort of women born in 1989 (relative risk 0.76, 95% confidence interval [0.68; 0.86]). Larger effects were found in later birth cohorts, in which vaccination was more widespread. CONCLUSION: Eleven years after the introduction of HPV vaccination, a drop in the incidence of cervical cancer was observable at the population level in the birth cohorts eligible for vaccination.

Colorectal Cancer Survival in German-Danish Border Regions-A Registry-Based Cohort Study.

The aim of this study was (i) to update the reporting of colorectal cancer survival differences over time in the German-Danish border region (Schleswig-Holstein, Southern Denmark, and Zealand) and (ii) to assess the extent to which it can be explained by stage and primary treatment. Incident invasive colorectal cancer cases diagnosed from 2004 to 2016 with a follow-up of vital status through 31 December 2017 were extracted from cancer registries. Analyses were conducted by anatomical subsite and for four consecutive periods. Kaplan-Meier curves and log-rank tests were computed. Cox regression models using data from Schleswig-Holstein from 2004 to 2007 as the reference category were run while controlling for age, sex, stage, and treatment. The cox regression models showed decreasing hazard ratios of death for all three regions over time for both anatomical subsites. The improvement was stronger in the Danish regions, and adjustment for age, sex, stage, and treatment attenuated the results only slightly. In 2014-2016, colon cancer survival was similar across regions, while rectal cancer survival was significantly superior in the Danish regions. Regional survival differences can only partially be explained by differing stage distribution and treatment and may be linked additionally to healthcare system reforms and screening efforts.

Survival of breast cancer patients in German-Danish border regions - A registry-based cohort study.

BACKGROUND AND AIM: Denmark reports slightly lower breast cancer survival before 2010 than its neighbouring country Germany. Previous research is limited by lacking stage and treatment information. This study aims to investigate differences in breast cancer survival between the bordering regions Schleswig-Holstein (Germany), Southern Denmark and Zealand (Denmark) using registry data including stage and treatment information. METHOD: Invasive female breast cancer cases diagnosed during 2004-2013 with follow up through 31st December 2014 were extracted from cancer registries. Cases notified by death certificates only and those aged 90+ years were excluded. Kaplan-Meier curves and log-rank tests were computed. Cox regression analysis was conducted with adjustment for year of diagnosis, age, stage, and treatment. RESULTS: The analytical sample included 42,966 cases. Kaplan-Meier curves and log-rank tests show significant survival differences between the regions. The Cox regression model adjusted for year of diagnosis and age shows significantly worse overall survival of breast cancer patients in both Danish regions compared to Schleswig-Holstein with hazard ratios (HR) of 1.09 (95 % CI: 1.04; 1.15) for patients from Southern Denmark (SD) and 1.25 (95 % CI: 1.18; 1.32) for residents of Zealand (ZL). This effect diminished after adjustment for stage and treatment (HR: 1.05 (SD), 1.09 (ZL) 95 % CI: 0.99; 1.10 (SD), 1.03; 1.15 (ZL)). CONCLUSION: Survival differences can be explained by differing stage distribution and treatment administration, which formerly were more favourable in Schleswig-Holstein. The survival gap will probably close due to Denmark's national screening program and increased use of adjuvant cancer therapy.

The acceptance and applicability of a patient-reported experience measurement tool in oncological care: a descriptive feasibility study in northern Germany.

BACKGROUND: Patient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs) are tools for assessing outcomes of and experiences with health care from the patient's perspective. In Germany, PROMs are widely used in research for evaluating patient outcomes and quality of care. However, the application of PREMs is rather scant, especially in oncology. The study aimed to assess the feasibility of patient-centred quality evaluation in oncological care in Germany using the German adaptation of the Danish National Cancer Patient Questionnaire. This questionnaire is a PREM/PROM-tool addressing patients of all cancer sites and covering the entire cancer patient pathway. METHODS: The Danish National Cancer Patient Questionnaire was translated into German via forward-backward translation. Face-validity was tested among three cancer patients in a conventional pre-test. The German adaptation contains 99 questions. A pilot test was carried out among 245 newly diagnosed breast and colorectal cancer patients in the German federal state Schleswig-Holstein. Patients were recruited via clinics participating in the Oncological Care Registry (12 specialised units in seven hospitals) and contacted six to nine months after diagnosis. Response behaviour and response patterns were compared to the Danish study population (n = 1964). RESULTS: The willingness among clinicians to support patient recruitment as well as the response rate of patients to the questionnaire was high (65%). Moreover, response behaviour and response patterns of German and Danish patients were consistent. Despite the generally good response behaviour of patients to the single items, the authors observed that questions assessing the diagnostic process did not fully capture German pathways. Only 19.3% of the German patients stated that their diagnostic process was initiated by a visit to a general practitioner (GP) in contrast to 52% in Denmark. The assessment of patient experiences in the diagnostic phase heavily focuses on experiences in general practice, which does not seem appropriate in the German health care setting. CONCLUSION: The translation was successful, and the feasibility of a future large-scale study within existing structures is given. However, some modifications of questions heavily related to the Danish health care system, especially referring to the diagnostic phase, are necessary.

[Baseline Incidence of HPV-Associated Invasive and in Situ Cancers in Germany Prior to Potential Vaccination Effects]. / Basisinzidenz HPV-assoziierter invasiver und in situ Karzinome in Deutschland vor Eintreten möglicher Impfeffekte.

AIM: Vaccines against human papillomaviruses are covered by statutory health insurances in Germany since March 2007. A reduction of cervical and other HPV-associated cancers is expected. Incidence rates of HPV-associated cancers preceding the introduction of vaccines are unknown. These, however, are important for assessing the effects of HPV vaccines. The aim of the present study was to bridge this knowledge gap. METHODS: Based on cancer registry data from seven federal states, average yearly incidence rates for cervical, anogenital, oropharyngeal cancers and cancers of the oral cavity were calculated for the period 2003 to 2007. Additionally, estimated numbers of cases give an overview on the total burden of disease caused by HPV-associated cancers in Germany. Incidence trends from 2003 to 2013 are also presented. RESULTS: In Germany, 11,686 cases of invasive HPV-associated cancers occurred yearly before the introduction of HPV vaccines. The most common sites are cervical cancers for women (4,387 cases) and oropharyngeal cancers for men (2,770 cases) with age-standardized incidence rates of 8.8 and 5.9 cases per 100,000 persons, respectively. The age-standardized incidence rate for in situ cervical cancers amounts to 23.5 cases per 100 000 women. Incidence trends show significant increases of invasive vulvar, penile and anorectal cancers. CONCLUSION: This work presents robust incidence data of HPV-associated cancers in Germany prior to the introduction of the HPV vaccine. Effects of HPV vaccines on the incidence of HPV-associated cancers can be estimated for Germany earliest in 2019.

Brief Report: Sexual Orientation in Individuals with Autistic Traits: Population Based Study of 47,000 Adults in Stockholm County.

We examined the association between autistic traits and sexual orientation in a general adult population (N = 47,356). Autistic traits were measured with the ten items Autistic Quotient questionnaire using a cut-off score of ≥ 6. Sexual orientation was assessed by self-report. Multinomial logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for sexual orientation categories. Participants with autistic traits were more likely to identify as bisexual (OR 1.73; 95% CI 1.01-2.9) and to feel that their sexual orientation could neither be described as hetero-, homo- nor bisexual (OR 3.05; 95% CI 2.56-3.63), compared to individuals without autistic traits. Autistic traits are associated with minority sexual orientation, and perhaps with uncertain self-identification and/or a defiance of traditional ways of categorizing sexual identity.

Involvement of the DNA mismatch repair system in cisplatin sensitivity of testicular germ cell tumours.

BACKGROUND: Testicular germ cell tumours (TGCT) are highly sensitive to cisplatin-based chemotherapy, but patients with tumours containing differentiated teratoma components are less responsive to this treatment. The cisplatin sensitivity in TGCT has previously been linked to the embryonic phenotype in the majority of tumours, although the underlying mechanism largely remains to be elucidated. The aim of this study was to investigate the role of the DNA mismatch repair (MMR) system in the cisplatin sensitivity of TGCT. METHODS: The expression pattern of key MMR proteins, including MSH2, MSH6, MLH1 and PMS2, were investigated during testis development and in the pathogenesis of TGCT, including germ cell neoplasia in situ (GCNIS). The TGCT-derived cell line NTera2 was differentiated using retinoic acid (10 µM, 6 days) after which MMR protein expression and activity, as well as cisplatin sensitivity, were investigated in both undifferentiated and differentiated cells. Finally, the expression of MSH2 was knocked down by siRNA in NTera2 cells after which the effect on cisplatin sensitivity was examined. RESULTS: MMR proteins were expressed in proliferating cells in the testes, while in malignant germ cells MMR protein expression was found to coincide with the expression of the pluripotency factor OCT4, with no or low expression in the more differentiated yolk sac tumours, choriocarcinomas and teratomas. In differentiated NTera2 cells we found a significantly (p < 0.05) lower expression of the MMR and pluripotency factors, as well as a reduced MMR activity and cisplatin sensitivity, compared to undifferentiated NTera2 cells. Also, we found that partial knockdown of MSH2 expression in undifferentiated NTera2 cells resulted in a significantly (p < 0.001) reduced cisplatin sensitivity. CONCLUSION: This study reports, for the first time, expression of the MMR system in fetal gonocytes, from which GCNIS cells are derived. Our findings in primary TGCT specimens and TGCT-derived cells suggest that a reduced sensitivity to cisplatin in differentiated TGCT components could result from a reduced expression of MMR proteins, in particular MSH2 and MLH1, which are involved in the recognition of cisplatin adducts and in activation of the DNA damage response pathway to initiate apoptosis.

Incidence trends of nonmelanoma skin cancer in Germany from 1998 to 2010.

BACKGROUND AND OBJECTIVES: Nonmelanoma skin cancer (NMSC) is the most common malignant neoplasm in Germany. However, little is known about incidence trends of NMSC and its main subtypes basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) in Germany. MATERIAL AND METHODS: Cancer registry data from fourteen German federal states was used to estimate age-standardized incidence rates by histologic subtype. Annual percentage changes (APC) were estimated in joinpoint regression models, in order to assess trend shifts in the years from 1998 to 2010. RESULTS: In Germany, incidence rates of NMSC showed a significant increase from 43.1 cases/100 000 in 1998 to 105.2 cases/100 000 in 2010. Incidence rates and incidence increases revealed large regional variations. Basal cell carcinoma was the most common tumor followed by SCC. Men were more frequently affected than women, but incidence increases were steeper in women. CONCLUSIONS: Results are consistent with national and international observations. The heterogeneity of incidence rates and their changes among federal states indicate that incidence changes are most likely related to improved case registration practices in German cancer registries.