Excess all-cause mortality in the evaluation of a screening trial to account for selective participation

All-Cause Mortality Risk in National Prostate Cancer Cohort: An Impact of Population-Based Prostate Cancer Screening

The aim of this study is to evaluate all-cause mortality risk differences before and during prostate cancer screening, with a profound focus on the differences between screened and not-screened patient groups. Prostate cancer cases diagnosed between 1998 and 2016 were identified from the population-based Lithuanian Cancer Registry and linked with screening status in the National Health Insurance Fund database. The analysis was stratified by a period of diagnosis and screening status. Standardized mortality ratios (SMRs) were used to assess all-cause and cause-specific mortality risk. The SMRs were calculated by dividing the observed number of deaths among prostate cancer patients by the expected number of deaths from the general population. All-cause SMR (1.45 (95% CI 1.42-1.48)) in the pre-screening period was higher compared to the screening period (SMR = 1.17 (95% CI 1.15-1.19)). An increased all-cause mortality risk among prostate cancer patients was observed in the not-screened patient population (SMR = 1.76 (95% CI 1.71-1.82)), while all-cause mortality risk in the screened patient population was similar to the general population (SMR = 1.00 (95% CI 0.97-1.02)). Screened patients with localized stage of disease had lower all-cause mortality risk than the general population (SMR = 0.72 (95% CI 0.70-0.75)). In conclusion, men with prostate cancer in Lithuania had excess all-cause mortality risk compared to the general population. The all-cause mortality risk among screened patients was not higher than expected.

Impact of cause of death adjudication on the results of the European prostate cancer screening trial

BACKGROUND

The European Randomised Study of Prostate Cancer Screening has shown a 21% relative reduction in prostate cancer mortality at 13 years. The causes of death can be misattributed, particularly in elderly men with multiple comorbidities, and therefore accurate assessment of the underlying cause of death is crucial for valid results. To address potential unreliability of end-point assessment, and its possible impact on mortality results, we analysed the study outcome adjudication data in six countries.

METHODS

Latent class statistical models were formulated to compare the accuracy of individual adjudicators, and to assess whether accuracy differed between the trial arms. We used the model to assess whether correcting for adjudication inaccuracies might modify the study results.

RESULTS

There was some heterogeneity in adjudication accuracy of causes of death, but no consistent differential accuracy by trial arm. Correcting the estimated screening effect for misclassification did not alter the estimated mortality effect of screening.

CONCLUSIONS

Our findings were consistent with earlier reports on the European screening trial. Observer variation, while demonstrably present, is unlikely to have materially biased the main study results. A bias in assigning causes of death that might have explained the mortality reduction by screening can be effectively ruled out.

Characteristics of men responding to an invitation to undergo testing for prostate cancer as part of a randomised trial

BACKGROUND

Sociodemographic characteristics are associated with participating in cancer screening and trials. We compared the characteristics of those responding with those not responding to a single invitation for prostate-specific antigen (PSA) testing for prostate cancer as part of the Cluster randomised triAl of PSA testing for Prostate cancer (CAP).

METHODS

Age, rurality and deprivation among 197,763 men from 271 cluster-randomised primary care centres in the UK were compared between those responding (n = 90,300) and those not responding (n = 100,953) to a prostate cancer testing invitation.

RESULTS

There was little difference in age between responders and nonresponders. Responders were slightly more likely to come from urban rather than rural areas and were slightly less deprived than those who did not respond.

CONCLUSION

These data indicate similarities in age and only minor differences in deprivation and urban location between responders and nonresponders. These differences were smaller, but in the same direction as those observed in other screening trials.

TRIAL REGISTRATION

ISRCTN92187251 . Registered on 29 November 2004.

Metastatic Prostate Cancer Incidence and Prostate-specific Antigen Testing: New Insights from the European Randomized Study of Screening for Prostate Cancer

BACKGROUND

The European Randomized Study of Screening for Prostate Cancer (ERSPC) has shown a 21% reduction in prostate cancer (PCa) mortality and a 1.6-fold increase in PCa incidence with prostate-specific antigen (PSA)-based screening (at 13 yr of follow-up). We evaluated PCa incidence by risk category at diagnosis across the study arms to assess the potential impact on PCa mortality.

DESIGN, SETTING, AND PARTICIPANTS

Information on arm, centre, T and M stage, Gleason score, serum PSA at diagnosis, age at randomisation, follow-up time, and vital status were extracted from the ERSPC database. Four risk categories at diagnosis were defined: 1, low; 2, intermediate; 3, high; 4, metastatic disease. PSA (≤100 or >100 ng/ml) was used as the indicator of metastasis.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Incidence rate ratios (IRRs) for screening versus control arm by risk category at diagnosis and follow-up time were calculated using Poisson regression analysis for seven centres. Follow-up was truncated at 13 yr. Missing data were imputed using chained equations. The analyses were carried out on an intention-to-treat basis.

RESULTS AND LIMITATIONS

In the screening arm, 7408 PCa cases were diagnosed and 6107 in the control arm. The proportion of missing stage, Gleason score, or PSA value was comparable in the two arms (8% vs 10%), but differed among centres. The IRRs were elevated in the screening arm for the low-risk (IRR: 2.14; 95% CI, 2.03-2.25) and intermediate-risk (IRR: 1.24; 95% CI, 1.16-1.34) categories at diagnosis, equal to unity for the high-risk category at diagnosis (IRR: 1.00; 95% CI, 0.89-1.13), and reduced for metastatic disease at diagnosis (IRR: 0.60; 95% CI, 0.52-0.70). The IRR of metastatic disease had temporal pattern similar to mortality, shifted forwards an average of almost 3 yr, although the mortality reduction was smaller.

CONCLUSIONS

The results confirm a reduction in metastatic disease at diagnosis in the screening arm, preceding mortality reduction by almost 3 yr.

PATIENT SUMMARY

The findings of this study indicate that the decrease in metastatic disease at diagnosis is the major determinant of the prostate cancer mortality reduction in the European Randomized study of Screening for Prostate Cancer.

Impact of Screening Program on Incidence of Colorectal Cancer: A Cohort Study in Italy

OBJECTIVES

Colorectal cancer (CRC) screening using the fecal occult blood test (FOBT) has been shown to be effective in reducing cause-specific mortality. However, although it detects pre-cancerous adenomas, it is uncertain whether FOBT reduces the incidence of invasive cancer. The objective is to evaluate the impact of screening with immunochemical FOBT (FIT) on CRC incidence and mortality.

METHODS

An organized screening program was implemented in 2005 in the province of Reggio Emilia (Northern Italy). The program invites the resident population aged 50-69 for FIT every 2 years. Subjects who test positive are referred for colonoscopy. Incidence was studied through cancer registry. Person-times of people aged 50-74 from 1997 to 2012 were classified for exposure to screening according to age and period. Furthermore, two open cohorts-one never screened (aged 50-69 in 1997) and one invited for screening (aged 50-69 in 2005)-were followed up for 8 years.

RESULTS

A total of 171,785 people have been invited, and approximately 70% have undergone FIT at least once (272,197 tests). The rate of colonoscopy participation has been about 90%, and 2896 cancers have been recorded (1237 in the screening period). The age-adjusted and sex-adjusted incidence rate ratios as compared with pre-screening were 1.60 (95% confidence interval (CI), 1.43-1.79), 0.86 (95% CI, 0.78-0.94), and 0.59 (95% CI, 0.50-0.69) for the first round, subsequent rounds, and post screening, respectively. Cumulative incidence and incidence-based mortality decreased by 10% (95% CI, 3-17%) and 27% (95% CI, 15-37%), respectively.

CONCLUSIONS

FIT screening leads to a decrease in the incidence of CRC and in its mortality.

Correction factors for self-selection when evaluating screening programmes

Objective

In screening programmes there is recognized bias introduced through participant self-selection (the healthy screenee bias). Methods used to evaluate screening programmes include Intention-to-screen, per-protocol, and the “post hoc” approach in which, after introducing screening for everyone, the only evaluation option is participants versus non-participants. All methods are prone to bias through self-selection. We present an overview of approaches to correct for this bias.

Methods

We considered four methods to quantify and correct for self-selection bias. Simple calculations revealed that these corrections are actually all identical, and can be converted into each other. Based on this, correction factors for further situations and measures were derived. The application of these correction factors requires a number of assumptions.

Results

Using as an example the German Neuroblastoma Screening Study, no relevant reduction in mortality or stage 4 incidence due to screening was observed. The largest bias (in favour of screening) was observed when comparing participants with non-participants.

Conclusions

Correcting for bias is particularly necessary when using the post hoc evaluation approach, however, in this situation not all required data are available. External data or further assumptions may be required for estimation.

Impact of Early Diagnosis of Prostate Cancer on Survival Outcomes

CONTEXT

The relationship between early detection of prostate cancer (PCa) and disease-specific mortality is still the subject of much debate.

OBJECTIVE

This review describes developments in PCa mortality rates and disease-stage shift on a population level. The main findings from the randomised screening trials are also discussed. Finally, we consider the expected consequences for the individual man interested in screening.

EVIDENCE ACQUISITION

The PubMed database was searched for trials of screening for PCa from inception through October 11, 2014. Supplementary information was collected by cross-referencing the reference lists.

EVIDENCE SYNTHESIS

Since the introduction of prostate-specific antigen testing, PCa incidence has risen, and a stage shift towards more favourable disease at diagnosis has been observed. PCa mortality rates are gradually decreasing. Although screening trials show conflicting results, the largest randomised trial of screening for PCa shows a 21% decrease in PCa-specific mortality. After correction for noncompliance and contamination, a risk reduction in PCa-specific mortality of up to 49% has been reported. The main side effect of screening is that some studies have estimated that approximately 50% of detected cases may represent overdiagnosis, which may be reduced by stopping screening in older men and using an individual risk-based approach.

CONCLUSIONS

To maximise the benefits while minimising the risk of overdiagnosis, future screening should follow an individual risk-based approach.

PATIENT SUMMARY

On a population level, the introduction of screening for prostate cancer (PCa) is associated with more men diagnosed but with more favourable disease. The largest screening study confirmed the reduction in death due to PCa. Individual risk estimation is important to best balance the benefits and potential harms of early detection.

A different method of evaluation of the ERSPC trial confirms that prostate-specific antigen testing has a significant impact on prostate cancer mortality

The advantages and disadvantages of two different methods of analyzing the European Randomized Study of Screening for Prostate Cancer (ERSPC) trial with respect to the effect of prostate-specific antigen (PSA) screening on prostate cancer (PCa) mortality (ie, disease-specific mortality analysis and excess mortality analysis) are discussed in depth. The traditional disease-specific mortality is the best end point, but it could be biased by misclassification of causes of death, and it does not take into account the possible effect of the screening process on other causes of death. Excess mortality analysis overcomes these problems, but the results could be biased if the expected mortality is not corrected for attendance status. Both methods, when applied to the ERSPC trials, demonstrate that no increase in non-PCa mortality occurred in the screening group and confirm that PSA screening decreases PCa mortality.