Validation and public health modelling of risk prediction models for kidney cancer using the UK Biobank

Development and external validation of a head and neck cancer risk prediction model

BACKGROUND

Head and neck cancer (HNC) incidence is on the rise, often diagnosed at late stage and associated with poor prognoses. Risk prediction tools have a potential role in prevention and early detection.

METHODS

The IARC-ARCAGE European case-control study was used as the model development dataset. A clinical HNC risk prediction model using behavioral and demographic predictors was developed via multivariable logistic regression analyses. The model was then externally validated in the UK Biobank cohort. Model performance was tested using discrimination and calibration metrics.

RESULTS

1926 HNC cases and 2043 controls were used for the development of the model. The development dataset model including sociodemographic, smoking, and alcohol variables had moderate discrimination, with an area under curve (AUC) value of 0.75 (95% CI, 0.74-0.77); the calibration slope (0.75) and tests were suggestive of good calibration. 384 616 UK Biobank participants (with 1177 HNC cases) were available for external validation of the model. Upon external validation, the model had an AUC of 0.62 (95% CI, 0.61-0.64).

CONCLUSION

We developed and externally validated a HNC risk prediction model using the ARCAGE and UK Biobank studies, respectively. This model had moderate performance in the development population and acceptable performance in the validation dataset. Demographics and risk behaviors are strong predictors of HNC, and this model may be a helpful tool in primary dental care settings to promote prevention and determine recall intervals for dental examination. Future addition of HPV serology or genetic factors could further enhance individual risk prediction.

Comparison of prediction models for cardiovascular and mortality risk in people with type 2 diabetes: An external validation in 23 685 adults included in the UK Biobank

AIMS

To validate cardiovascular risk prediction models for individuals with diabetes using the UK Biobank in order to assess their applicability.

METHODS

We externally validated 19 cardiovascular risk scores from seven risk prediction models (Chang et al., Framingham, University of Hong Kong-Singapore [HKU-SG], Li et al, RECODe [risk equations for complications of type 2 diabetes], SCORE [Systematic Coronary Risk Evaluation] and the UK Prospective Diabetes Study Outcomes Model 2 [UKPDS OM2]), identified from systematic reviews, using UK Biobank data from 2006 to 2021 (n = 23 685; participant age 40-71 years, 63.5% male). We evaluated performance by assessing the discrimination and calibration of the models for the endpoints of mortality, cardiovascular mortality, congestive heart failure, myocardial infarction, stroke, and ischaemic heart disease.

RESULTS

Over a total of 269 430 person-years of follow-up (median 11.89 years), the models showed low-to-moderate discrimination performance on external validation (concordance indices [c-indices] 0.50-0.71). Most models had low calibration with overprediction of the observed risk. RECODe outperformed other models across four comparable endpoints for discrimination: all-cause mortality (c-index 0.67, 95% confidence interval [CI] 0.65-0.69), congestive heart failure (c-index 0.71, 95% CI 0.69-0.72), myocardial infarction (c-index 0.67, 95% CI 0.65-0.68); and stroke (c-index 0.65, 95% CI 0.62-0.68), and for calibration (except for all-cause mortality). The UKPDS OM2 had comparable performance to RECODe for all-cause mortality (c-index 0.67, 95% CI 0.66-0.69) and cardiovascular mortality (c-index 0.71, 95% CI 0.70-0.73), but worse performance for other outcomes. The models performed better for younger participants and somewhat better for non-White ethnicities. Models developed from non-Western datasets showed worse performance in our UK-based validation set.

CONCLUSIONS

The RECODe model led to better risk estimations in this predominantly White European population. Further validation is needed in non-Western populations to assess generalizability to other populations.

Effect of smoking, hypertension and lifestyle factors on kidney cancer - perspectives for prevention and screening programmes

Renal cell carcinoma (RCC) incidence has doubled over the past few decades. However, death rates have remained stable as the number of incidental renal mass diagnoses peaked. RCC has been recognized as a European health care issue, but to date, no screening programmes have been introduced. Well-known modifiable risk factors for RCC are smoking, obesity and hypertension. A direct association between cigarette consumption and increased RCC incidence and RCC-related death has been reported, but the underlying mechanistic pathways for this association are still unclear. Obesity is associated with an increased risk of RCC, but interestingly, improved survival outcomes have been reported in obese patients, a phenomenon known as the obesity paradox. Data on the association between other modifiable risk factors such as diet, dyslipidaemia and physical activity with RCC incidence are conflicting, and potential mechanisms underlying these associations remain to be elucidated.

Estimating the Effectiveness of Kidney Cancer Screening Within Lung Cancer Screening Programmes: A Validation in UK Biobank

In the absence of population-based screening, addition of screening for kidney cancer to lung cancer screening could provide an efficient and low-resource means to improve early detection. In this study, we used the UK Biobank cohort (n = 442 865) to determine the performance of the Yorkshire Lung Cancer Screening Trial (YLST) eligibility criteria for selecting individuals for kidney cancer screening. We measured the performance of two models widely used to determine eligibility for lung cancer screening (PLCO[m2012] and the Liverpool-Lung-Project-v2) and the performance of the combined YLST criteria. We found that the lung cancer models have discrimination (area under the receiver operating curve) between 0.60 and 0.68 for kidney cancer. In the UK, one in four cases (25%) of kidney cancer cases is expected to occur in those eligible for lung cancer screening, and one case of kidney cancer detected for every 200 people invited to lung cancer screening. These results suggest that adding kidney cancer screening to lung cancer screening would be an effective strategy to improve early detection rates of kidney cancer. However, most kidney cancers would not be picked up by this approach. This analysis does not address other important considerations about kidney cancer screening, such as overdiagnosis. PATIENT SUMMARY: It has been proposed that adding-on kidney cancer screening to lung cancer screening (both carried out by a computed tomography scan of the chest/abdomen) would be an easy and low-cost way of detecting cases of kidney cancer earlier, when these can be treated more easily. Lung cancer screening is usually targeted at people who are at a high risk (eg, older smokers); therefore, here we look at whether the same group of people are also at a high risk of kidney cancer. Our analysis shows that one in four people later diagnosed with kidney cancer are also at a high risk of lung cancer; hence, a combined screening programme could detect up to a quarter of kidney cancers.

The current state of genetic risk models for the development of kidney cancer: a review and validation

OBJECTIVE

To review the current state of genetic risk models for predicting the development of kidney cancer, by identifying and comparing the performance of published models.

METHODS

Risk models were identified from a recent systematic review and the Cancer-PRS web directory. A narrative synthesis of the models, previous validation studies and related genome-wide association studies (GWAS) was carried out. The discrimination and calibration of the identified models was then assessed and compared in the UK Biobank (UKB) cohort (cases, 452; controls, 487 925).

RESULTS

A total of 39 genetic models predicting the development of kidney cancer were identified and 31 were validated in the UKB. Several of the genetic-only models (seven of 25) and most of the mixed genetic-phenotypic models (five of six) had some discriminatory ability (area under the receiver operating characteristic curve >0.5) in this cohort. In general, models containing a larger number of genetic variants identified in GWAS performed better than models containing a small number of variants associated with known causal pathways. However, the performance of the included models was consistently poorer than genetic risk models for other cancers.

CONCLUSIONS

Although there is potential for genetic models to identify those at highest risk of developing kidney cancer, their performance is poorer than the best genetic risk models for other cancers. This may be due to the comparatively small number of genetic variants associated with kidney cancer identified in GWAS to date. The development of improved genetic risk models for kidney cancer is dependent on the identification of more variants associated with this disease. Whether these will have utility within future kidney cancer screening pathways is yet to determined.