Cohort profile: radiotherapy dataset (RTDS) in England

Emulating an existing trial of treatments for prostate cancer using real-world data: challenges and lessons learned

OBJECTIVES

If randomized controlled trials can be successfully emulated using real-world data (RWD), confidence in the validity of RWD for estimating treatment effects for questions that have not been assessed in trials increases. We used routinely collected administrative and clinical national linked datasets from England to emulate the PR07 trial for high-risk prostate cancer patients, which compared the effects of radiotherapy added to hormone therapy (RT+HT) within 8 weeks of randomization (the "grace period") and hormone therapy (HT) only on all-cause mortality. We highlight methodological choices required and challenges encountered in emulating this trial.

STUDY DESIGN AND SETTING

Patients diagnosed with prostate cancer from 2014 to 2020 were identified from the routine national linked datasets. Diagnosis was taken as the time zero. As few patients initiated radiotherapy within 8 weeks of diagnosis, we considered target trials with grace periods of 4-6 months. Estimands of interest were hazard ratios (HRs) and survival probabilities over 7 years. The cloning-censoring-and-weighting (CCW) approach was used to control for measured confounding and to allow for the grace period. We also used an extension (the "landmark-CCW" approach), in which we consider several time-origins post-diagnosis, enabling us to use a grace period of 8 weeks as in PR07.

RESULTS

A total of 2,690 patients were eligible for inclusion in the emulated trial. The CCW analysis using a grace period of 6 months gave an estimated HR of 0.48 (95% confidence interval [CI]: 0.34-0.60) and 7-year survival estimates of 80.7% (95% CI: 74.3-87.0) for the RT+HT strategy and 65.6% (95% CI: 62.8-68.1) for HT only strategy, and corresponding risk difference of 15.1% (95% CI: 11.5-18.9). The corresponding HR from the landmark-CCW approach was 0.58 (95% CI: 0.51-0.65) and with survival estimates of 80.7% (95% CI: 77.7-83.8) for RT+HT strategy and 69.8% (95% CI: 68.2-71.4) for the HT only strategy, and a risk difference of 10.9% (95% CI: 6.3-15.9).

CONCLUSION

Our findings from the emulated trial using RWD are broadly consistent with those from PR07, with RT+HT estimated to result in better survival compared to HT only. However, the findings were not replicated exactly, with PR07 reporting an HR of 0.77 (95% CI: 0.61-0.98) over 7 years of follow-up. Differences may be in part due to challenges in defining time zero and allowing for a treatment grace period of the same duration as in PR07. Our study considered ways in which these challenges can be addressed, and our findings affirm the utility of RWD for estimating treatment effects.

PLAIN LANGUAGE SUMMARY

Clinical trials are the best way to test whether treatments work, but they are expensive, take years to complete, and focus on narrow research questions. If we can use real-world data (RWD) such as patient health records to mimic these trials, we may be able to answer additional medical questions relevant to patients who are prescribed these medications. This study aimed to see if we could recreate the results of a past clinical trial (PR07) using national health data from England. The PR07 trial looked at two treatments for high-risk prostate cancer: hormone therapy (HT) alone and radiotherapy added to hormone therapy (RT+HT) within 8 weeks of starting the study. This trial was chosen for several reasons. It included high-risk patients who were studied for up to 7 years, meaning that there was enough information to study survival outcomes. Being a major UK-based trial, it was useful for comparing with the data recorded in UK national health data. The trial also allowed some flexibility in when treatment started, which was an interesting factor to examine. Its main goal was to see if adding radiotherapy to hormone therapy provided extra benefits to patients. We wanted to see whether the trial results were replicated using the UK health data. If results were replicated, it would provide confidence in further exploration of questions not covered by the trial. In a trial, randomization time is the point where doctors allocate patients to one of the treatments being assessed, usually 1 new treatment and 1 control treatment. Patients in the study are then followed up from that point. However, defining a starting point in a non-trial setting, using UK national health data, is not as straightforward. In the PR07 trial, patients started RT+HT within 8 weeks of randomization. In the UK datasets, we use diagnosis as a starting point, because it is available for both treatment groups. However, very few patients started RT+HT within 8 weeks of diagnosis. To address this, we allowed for longer periods of 4-6 months from diagnosis for RT+HT initiation. Recent developments provided statistical methods for estimating the cause-effect relationship between treatment received and survival patterns. In this study, we show how these approaches can be used to estimate survival patterns if all patients had RT+HT within 4-6 months from diagnosis compared with if no patients had any RT within 4-6 months. We account for the different treatment initiation times and the main differences between the RT+HT and HT only patients. Using these methods, we estimate that RT+HT has an 11%-15% higher survival rate at 7 years compared to HT only. We discuss similarities and differences between these findings and those in the original PR07 trial.

Conducting a National RT-QI Project - Challenges and Opportunities

Over the past decade, there has been an increased interest in defining and monitoring quality indicators (QI) in the field of oncology including the field of radiation oncology. The comprehensive gathering and analysis of QIs on a multicentric scale offer valuable insights into identifying gaps in clinical practice and fostering continuous improvement. This article delineates the evolution and results of the Belgian national project dedicated to radiotherapy-specific QIs while also exploring the challenges and opportunities inherent in implementing such a multi-centric initiative.

An Overview of Real-World Data Infrastructure for Cancer Research

AIMS

There is increasing interest in the opportunities offered by Real World Data (RWD) to provide evidence where clinical trial data does not exist, but access to appropriate data sources is frequently cited as a barrier to RWD research. This paper discusses current RWD resources and how they can be accessed for cancer research.

MATERIALS AND METHODS

There has been significant progress on facilitating RWD access in the last few years across a range of scales, from local hospital research databases, through regional care records and national repositories, to the impact of federated learning approaches on internationally collaborative studies. We use a series of case studies, principally from the UK, to illustrate how RWD can be accessed for research and healthcare improvement at each of these scales.

RESULTS

For each example we discuss infrastructure and governance requirements with the aim of encouraging further work in this space that will help to fill evidence gaps in oncology.

CONCLUSION

There are challenges, but real-world data research across a range of scales is already a reality. Taking advantage of the current generation of data sources requires researchers to carefully define their research question and the scale at which it would be best addressed.

National epidemiology of digital papillary adenocarcinoma in England 2013-2020: a population-based registry study

BACKGROUND

Digital papillary adenocarcinoma (DPA), formerly known as aggressive DPA, is a rare adnexal cancer of sweat gland differentiation with metastatic potential. DPA epidemiology and patient outcome data are prerequisites for developing diagnostic and therapeutic guidance, which are lacking for this rare cancer.

OBJECTIVES

To report the incidence, patient demographics and treatment of patients with DPA in England from 1 January 2013 to 31 December 2020 using national cancer registry data.

METHODS

DPA diagnoses in England during 2013-2020 were identified from the National Cancer Registration and Analysis Service dataset using morphology and behaviour codes. These were registered from routinely collected pathology reports, along with cancer outcomes and services datasets. The 2013 European age-standardized incidence rates (EASRs) were calculated.

RESULTS

In total, 36 cases of DPA (7 in women and 29 in men) were diagnosed. The median age at diagnosis for the cohort was 54 years (interquartile range 46-64). The most frequently affected sites were the upper limbs (81%). All patients in the cohort received surgical excisions. The EASR was 0.10 (95% confidence interval 0.07-0.14) per 1 000 000 person-years.

CONCLUSIONS

This study reports the incidence and variation of DPA in England between 2013 and 2020. DPA was more common in older men and predominantly affected the upper limbs. This finding supports the need to develop a national policy for the reporting and management of DPA as well as development of a clinical guideline.

Risks of adverse obstetric outcomes among female survivors of adolescent and young adult cancer in England (TYACSS): a population-based, retrospective cohort study

BACKGROUND

There are limited data on the risks of obstetric complications among survivors of adolescent and young adult cancer with most previous studies only reporting risks for all types of cancers combined. The aim of this study was to quantify deficits in birth rates and risks of obstetric complications for female survivors of 17 specific types of adolescent and young adult cancer.

METHODS

The Teenage and Young Adult Cancer Survivor Study (TYACSS)-a retrospective, population-based cohort of 200 945 5-year survivors of cancer diagnosed at age 15-39 years from England and Wales-was linked to the English Hospital Episode Statistics (HES) database from April 1, 1997, to March 31, 2022. The cohort included 17 different types of adolescent and young adult cancers. We ascertained 27 specific obstetric complications through HES among 96 947 women in the TYACSS cohort. Observed and expected numbers for births and obstetric complications were compared between the study cohort and the general population of England to identify survivors of adolescent and young adult cancer at a heighted risk of birth deficits and obstetric complications relative to the general population.

FINDINGS

Between April 1, 1997, and March 31, 2022, 21 437 births were observed among 13 886 female survivors of adolescent and young adult cancer from England, which was lower than expected (observed-to-expected ratio: 0·68, 95% CI 0·67-0·69). Other survivors of genitourinary, cervical, and breast cancer had under 50% of expected births. Focusing on more common (observed ≥100) obstetric complications that were at least moderately in excess (observed-to-expected ratio ≥1·25), survivors of cervical cancer were at risk of malpresentation of fetus, obstructed labour, amniotic fluid and membranes disorders, premature rupture of membranes, preterm birth, placental disorders including placenta praevia, and antepartum haemorrhage. Survivors of leukaemia were at risk of preterm delivery, obstructed labour, postpartum haemorrhage, and retained placenta. Survivors of all other specific cancers had no more than two obstetric complications that exceeded an observed-to-expected ratio of 1·25 or greater.

INTERPRETATION

Survivors of cervical cancer and leukaemia are at risk of several serious obstetric complications; therefore, any pregnancy should be considered high-risk and would benefit from obstetrician-led antenatal care. Despite observing deficits in birth rates across all 17 different types of adolescent and young adult cancer, we provide reassurance for almost all survivors of adolescent and young adult cancer concerning their risk of almost all obstetric complications. Our results provide evidence for the development of clinical guidelines relating to counselling and surveillance of obstetrical risk for female survivors of adolescent and young adult cancer.

FUNDING

Children with Cancer UK, The Brain Tumour Charity, and Academy of Medical Sciences.

Cancer incidence, treatment, and survival in the prison population compared with the general population in England: a population-based, matched cohort study

BACKGROUND

The growing and ageing prison population in England makes accurate cancer data of increasing importance for prison health policies. This study aimed to compare cancer incidence, treatment, and survival between patients diagnosed in prison and the general population.

METHODS

In this population-based, matched cohort study, we used cancer registration data from the National Cancer Registration and Analysis Service in England to identify primary invasive cancers and cervical cancers in situ diagnosed in adults (aged ≥18 years) in the prison and general populations between Jan 1, 1998, and Dec 31, 2017. Ministry of Justice and Office for National Statistics population data for England were used to calculate age-standardised incidence rates (ASIR) per year and age-standardised incidence rate ratios (ASIRR) for the 20-year period. Patients diagnosed with primary invasive cancers (ie, excluding cervical cancers in situ) in prison between Jan 1, 2012, and Dec 31, 2017 were matched to individuals from the general population and linked to hospital and treatment datasets. Matching was done in a 1:5 ratio according to 5-year age group, gender, diagnosis year, cancer site, and disease stage. Our primary objectives were to compare the incidence of cancer (1998-2017); the receipt of treatment with curative intent (2012-17 matched cohort), using logistic regression adjusted for matching variables (excluding cancer site) and route to diagnosis; and overall survival following cancer diagnosis (2012-17 matched cohort), using a Cox proportional hazards model adjusted for matching variables (excluding cancer site) and route to diagnosis, with stratification for the receipt of any treatment with curative intent.

FINDINGS

We identified 2015 incident cancers among 1964 adults (1556 [77·2%] men and 459 [22·8%] women) in English prisons in the 20-year period up to Dec 31, 2017. The ASIR for cancer for men in prison was initially lower than for men in the general population (in 1998, ASIR 119·33 per 100 000 person-years [95% CI 48·59-219·16] vs 746·97 per 100 000 person-years [742·31-751·66]), but increased to a similar level towards the end of the study period (in 2017, 856·85 per 100 000 person-years [675·12-1060·44] vs 788·59 per 100 000 person-years [784·62-792·57]). For women, the invasive cancer incidence rate was low and so ASIR was not reported for this group. Over the 20-year period, the incidence of invasive cancer for men in prison increased (incidence rate ratio per year, 1·05 [95% CI 1·04-1·06], during 1999-2017 compared with 1998). ASIRRs showed that over the 20-year period, overall cancer incidence was lower in men in prison than in men in the general population (ASIRR 0·76 [95% CI 0·73-0·80]). The difference was not statistically significant for women (ASIRR 0·83 [0·68-1·00]). Between Jan 1, 2012, and Dec 31, 2017, patients diagnosed in prison were less likely to undergo curative treatment than matched patients in the general population (274 [32·3%] of 847 patients vs 1728 [41·5%] of 4165; adjusted odds ratio (OR) 0·72 [95% CI 0·60-0·85]). Being diagnosed in prison was associated with a significantly increased risk of death on adjustment for matching variables (347 deaths during 2021·9 person-years in the prison cohort vs 1626 deaths during 10 944·2 person-years in the general population; adjusted HR 1·16 [95% CI 1·03-1·30]); this association was partly explained by stratification by curative treatment and further adjustment for diagnosis route (adjusted HR 1·05 [0·93-1·18]).

INTERPRETATION

Cancer incidence increased in people in prisons in England between 1998 and 2017, with patients in prison less likely to receive curative treatments and having lower overall survival than the general population. The association with survival was partly explained by accounting for differences in receipt of curative treatment and adjustment for diagnosis route. Improved routine cancer surveillance is needed to inform prison cancer policies and decrease inequalities for this under-researched population.

FUNDING

UK National Institute for Health and Care Research, King's College London, and Strategic Priorities Fund 2019/20 of Research England via the University of Surrey.

The comprehensive English National Lynch Syndrome Registry: development and description of a new genomics data resource

Background

Lynch Syndrome (LS) is a cancer predisposition syndrome caused by constitutional pathogenic variants in the mismatch repair (MMR) genes. To date, fragmentation of clinical and genomic data has restricted understanding of national LS ascertainment and outcomes, and precluded evaluation of NICE guidance on testing and management. To address this, via collaboration between researchers, the National Disease Registration Service (NDRS), NHS Genomic Medicine Service Alliances (GMSAs), and NHS Regional Clinical Genetics Services, a comprehensive registry of LS carriers in England has been established.

Methods

For comprehensive ascertainment of retrospectively identified MMR pathogenic variant (PV) carriers (diagnosed prior to January 1, 2023), information was retrieved from all clinical genetics services across England, then restructured, amalgamated, and validated via a team of trained experts in NDRS. An online submission portal was established for prospective ascertainment from January 1, 2023. The resulting data, stored in a secure database in NDRS, were used to investigate the demographic and genetic characteristics of the cohort, censored at July 25, 2023. Cancer outcomes were investigated via linkage to the National Cancer Registration Dataset (NCRD).

Findings

A total of 11,722 retrospective and 570 prospective data submissions were received, resulting in a comprehensive English National Lynch Syndrome Registry (ENLSR) comprising 9030 unique individuals. The most frequently identified pathogenic MMR genes were MSH2 and MLH1 at 37.2% (n = 3362) and 29.1% (n = 2624), respectively. 35.9% (n = 3239) of the ENLSR cohort received their LS diagnosis before their first cancer diagnosis (presumptive predictive germline test). Of these, 6.3% (n = 204) developed colorectal cancer, at a median age of initial diagnosis of 51 (IQR 40-62), compared to 73 years (IQR 64-80) in the general population (p < 0.0001).

Interpretation

The ENLSR represents the first comprehensive national registry of PV carriers in England and one of the largest cohorts of MMR PV carriers worldwide. The establishment of a secure, centralised infrastructure and mechanism for routine registration of newly identified carriers ensures sustainability of the data resource.

Funding

This work was funded by the Wellcome Trust, Cancer Research UK and Bowel Cancer UK. The funder of this study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

Cohort study to assess geographical variation in cholangiocarcinoma treatment in England

BACKGROUND

Outcomes for cholangiocarcinoma (CCA) are extremely poor owing to the complexities in diagnosing and managing a rare disease with heterogenous sub-types. Beyond curative surgery, which is only an option for a minority of patients diagnosed at an early stage, few systemic therapy options are currently recommended to relieve symptoms and prolong life. Stent insertion to manage disease complications requires highly specialised expertise. Evidence is lacking as to how CCA patients are managed in a real-world setting and whether there is any variation in treatments received by CCA patients.

AIM

To assess geographic variation in treatments received amongst CCA patients in England.

METHODS

Data used in this cohort study were drawn from the National Cancer Registration Dataset (NCRD), Hospital Episode Statistics and the Systemic Anti-Cancer Therapy Dataset. A cohort of 8853 CCA patients diagnosed between 2014-2017 in the National Health Service in England was identified from the NCRD. Potentially curative surgery for all patients and systemic therapy and stent insertion for 7751 individuals who did not receive surgery were identified as three end-points of interest. Linear probability models assessed variation in each of the three treatment modalities according to Cancer Alliance of residence at diagnosis, and for socio-demographic and clinical characteristics at diagnosis.

RESULTS

Of 8853 CCA patients, 1102 (12.4%) received potentially curative surgery. The mean [95% confidence interval (CI)] percentage-point difference from the population average ranged from -3.96 (-6.34 to -1.59)% to 3.77 (0.54 to 6.99)% across Cancer Alliances in England after adjustment for patient sociodemographic and clinical characteristics, showing statistically significant variation. Amongst 7751 who did not receive surgery, 1542 (19.9%) received systemic therapy, with mean [95%CI] percentage-point difference from the population average between -3.84 (-8.04 to 0.35)% to 9.28 (1.76 to 16.80)% across Cancer Alliances after adjustment, again showing the presence of statistically significant variation for some regions. Stent insertion was received by 2156 (27.8%), with mean [95%CI] percentage-point difference from the population average between -10.54 (-12.88 to -8.20)% to 13.64 (9.22 to 18.06)% across Cancer Alliances after adjustment, showing wide and statistically significant variation from the population average. Half of 8853 patients (n = 4468) received no treatment with either surgery, systemic therapy or stent insertion.

CONCLUSION

Substantial regional variation in treatments received by CCA patients was observed in England. Such variation could be due to differences in case-mix, clinical practice or access to specialist expertise.