This study describes the epidemiology of cancer in 4033 incident kidney transplant recipients in Scotland, followed up for a median of over 6 years. Compared with the general population, there is an increased risk of cancer, in particular non-melanomatous skin cancer, lymphoma and kidney cancer. The increased risk was greatest in transplant recipients under 40 years, at 7 times that of the general population, whilst recipients aged 60 years and over had a cancer rate 3 times that of the general population. The incidence of the most common cancers in the general population, including lung, breast, colorectal, and prostate, were minimally or not increased post-transplant in this study.

Our findings align with previous studies identifying risk factors for post-transplant cancer, including older age at transplantation, male sex, longer duration of dialysis pre-transplant, and prior history of cancer. In 6000 transplant recipients in Australia and New Zealand, pre-transplant dialysis duration was an independent risk factor for cancers of the urinary tract and lung, highlighting a need for vigilance in individuals with greater dialysis vintage [19]. Similarly, a US study of over 200,000 transplant recipients reported skin cancer risk was associated with older age, male sex, White ethnicity and pre-transplant malignancy [20]. Exposure to UV radiation and susceptibility of keratinocytes are key factors in the development of non-melanomatous skin cancer, and likely explains the variation in risk by geographical location with skin cancers particularly prevalent in Australia and New Zealand where there is high UV exposure [21, 22]. It is important to note however that even in Scotland, where vitamin D deficiency due to lack of UV exposure is common [23], non-melanomatous skin cancers still comprised over 50% of all cancers. Individuals with a history of cancer are also at an increased risk of recurrence. A study from Italy found nearly 40% of patients with non-melanomatous skin cancer developed a second lesion, and patients with a primary non-cutaneous cancer had an increased risk of developing a second non-cutaneous cancer [24].

Previous studies have identified further risk factors for post-transplant cancer. In 7000 patients in Australia and New Zealand, recipients of extended criteria deceased donor kidneys had a 1.5 fold increased risk of cancer, particularly genitourinary cancer and lymphoma, compared to those receiving a live donor organ [25]. Immune suppression also impacts on cancer risk, with pre-transplant cyclophosphamide or rituximab for the treatment of glomerular diseases [26], and the use of lymphocyte depleting agents being identified as risk factors [27, 28]. Cancers with a viral aetiology are more frequently observed during periods with a functioning graft when immune suppression is greatest, whilst CKD-related cancers such as kidney cancers may have a higher incidence after transplant failure [29]. The role of viral infections in the aetiology of post-transplant cancer has been most clearly defined with Epstein-Barr virus, with recipient serostatus a major determinant of lymphoma risk [30]. BK virus may also be implicated in urothelial malignancy, with studies showing an increased rate of bladder cancer in patients with BK nephropathy or BK viraemia [31], and a fifth of bladder cancers in transplant recipients containing BK viral sequences [32].

Cancer-related mortality is greater in transplant recipients than the general population [33]. The stage of cancer at presentation is a key factor in outcomes but we found cancer stage at diagnosis similar to the general population, aside from bladder cancer which tended to present more advanced. This differs from a study of 635 transplant recipients from the Israel Penn registry, which found that aside from kidney cancer most de novo cancers were more advanced at presentation than the general population [34]. A study of over 2500 transplant recipients in Australia and New Zealand with de novo cancer found the proportion presenting with lymph node or metastatic spread was approximately 20% for breast, 45% for lung, and 40% for colorectal cancer [35], similar to what is reported here. The high proportion of cases of advanced bladder cancer at presentation here merits discussion. As all transplant patients are informally screened with urinalysis at every clinic visit, it is a concern bladder cancer is not diagnosed early. This may reflect complacency regarding non-visible haematuria, which could also represent recurrent renal disease, or the disease being more aggressive in transplant recipients.

International guidelines recommend transplant recipients undergo the same solid organ cancer screening as the general population, though different diagnostic sensitivities, increased competing risks for deaths e.g. from cardiovascular disease, and reduced life expectancy in transplant recipients may impact screening effectiveness [36,37,38]. In Scotland, national screening programmes exist for cervical (cervical smear every 5 years for women aged 25–64 years), breast (mammogram every 3 years for women aged 50–70 years), and colorectal cancer (faecal occult blood or faecal immunochemical testing every 2 years between 50–74 years) [39]. The proportion of cancer cases picked up by screening in eligible transplant recipients was similar to that in the general population (around 30% of colorectal, 50% of breast and 40% of cervical cancers), although data on the uptake of screening among transplant patients compared to the general population was not available [15]. Screening for kidney cancer is not currently recommended, with a Markov model suggesting minimal survival benefit and an unfavourable cost-effectiveness ratio [40] which may relate to poor sensitivity of ultrasound in detecting lesions in abnormal end-stage kidneys [38]. The increased rate of kidney cancers in our cohort should however be noted, and a high index of suspicion should be maintained by transplant physicians. For skin cancers, guidance is for regular self-examination, and for a skin review biannually for the first 5 years post-transplant and annually thereafter [37]. A 2020 UK survey however found only 55% of transplant centres have a skin surveillance service, and of those that do two-thirds of assessments were performed by a non-skin cancer specialist [41]. Ensuring appropriate funding and training is available to facilitate such services and enable collaboration between skin cancer and transplant specialists is key to improving skin surveillance rates, which is vital given the high incidence rate of skin cancers in our population.

Whilst the increased risk of cancer post-transplant can be explained by immune suppression and susceptibility to viral-mediated cancers, ascertainment bias should also be considered. Kidney transplant recipients are closely monitored and frequently experience symptoms such as gastrointestinal disturbance relating to transplant medications. Such symptoms might prompt cross-sectional imaging and could explain the increased rate of urinary tract cancers, though no consistent increased rate of lung or colorectal cancer were seen in our population. Similarly, regular skin examinations provide opportunity for diagnosing skin cancers that may not have been detected until later in the disease course, when competing risks such as cardiovascular disease may have meant they would not have had a clinical impact on the patient.

This study has several strengths. We provide contemporary data with complete geographical coverage and a long duration of follow up for the population of Scotland, where there is universal access to health care. Results are generalisable to other high-income countries and provide vital information for patients and clinicians involved in transplant care. The examined registries have high accuracy and completeness rates [42], allow examination of cancer staging in transplant recipients and the general population, and enable comparison to other transplant populations across the world.

There are limitations to this study. Firstly, data on immune suppression were not available. Kidney transplantation in Scotland is performed in 2 centres with similar practice. Standard immune suppression prior to 2000 comprised methylprednisolone induction with prednisolone, ciclosporin and azathioprine maintenance. Basiliximab induction was introduced around 2000, and maintenance immune suppression changed to prednisolone, tacrolimus and mycophenolate mofetil between 2005–2007. Secondly, this study has not accounted for how time post-transplant impacts on the risk of cancer. This is likely to be non-linear i.e. there may be a cumulative risk relating to time on immune suppression that results in a greater risk with increasing time post-transplant, coupled with the lag time between exposure, cancer development and subsequent detection. Data were not available on other potential risk factors for cancer, such as smoking status, ethnicity, body mass index, sun exposure, donor type or viral mismatches. It is possible that some cancers, in particular non-melanomatous skin cancers, are under-reported in the Cancer Registry. This is because some may be treated in primary care without further records of the diagnosis, and others may not be brought to medical attention. Cancers however do not require pathology to be registered e.g. skin cancers treated with cryotherapy, and under-ascertainment of non-melanomatous skin cancers should not differentially affect transplant patients or lead to artefactual differences in apparent incidence compared to the general population. Finally, our data span the era of COVID-19 which may impact on incidence rates, as observed in the general population [43].

In conclusion, this study shows the rate of cancer in kidney transplant recipients is equivalent to someone 20 years older in the general population. The most frequently observed cancers were non-melanomatous skin cancers, lymphoma, and kidney cancer, reflecting a different distribution to the general population. The risk of cancer differs with age, sex, prior duration of dialysis, and pre-transplant malignancy, with younger transplant recipients having a disproportionate increased relative risk of cancer. These findings can guide patient counselling prior to transplantation, emphasise the importance of engaging transplant recipients in cancer surveillance, and highlight the need for robust pathways for diagnosing and treating cancer in kidney transplant recipients.