Assessment of kidney transplant suitability for patients with prior cancers: is it time for a rethink?

Summary

Kidney transplant recipients have up to a 100‐fold greater risk of incident cancer compared with the age/sex‐matched general population, attributed largely to chronic immunosuppression. In patients with a prior history of treated cancers, the type, stage and the potential for cancer recurrence post‐transplant of prior cancers are important factors when determining transplant suitability. Consequently, one of the predicaments facing transplant clinicians is to determine whether patients with prior cancers are eligible for transplantation, balancing between the accelerated risk of death on dialysis, the projected survival benefit and quality of life gains with transplantation, and the premature mortality associated with the potential risk of cancer recurrence post‐transplant. The guidelines informing transplant eligibility or screening and preventive strategies against cancer recurrence for patients with prior cancers are inconsistent, underpinned by uncertain evidence on the estimates of the incidence of cancer recurrence and the lack of stage‐specific outcomes data, particularly among those with multiple myeloma or immune‐driven malignancies such as melanomas. With the advent of newer anti‐cancer treatment options, it is unclear whether the current guidelines for those with prior cancers remain appropriate. This review will summarize the uncertainties of evidence informing the current recommendations regarding transplant eligibility of patients with prior cancers.

Introduction

Kidney transplantation is often the treatment of choice for a proportion of patients with end‐stage kidney disease (ESKD), conferring a significant survival advantage and improvement in quality of life compared to dialysis treatment 1, 2. Advances in the management of cardiovascular disease (CVD) have resulted in a substantial reduction in CVD mortality in kidney transplant recipients, but despite improved access to cancer screening and novel treatment options, cancer remains a major cause of morbidity and mortality in this population 3, 4. In Australia, cancer has surpassed CVD as the most frequent cause of mortality in kidney transplant recipients, with similar trends observed worldwide 4, 5. Even though cancer is a feared complication for kidney transplant recipients, many patients would still consider this as an acceptable risk in the trade‐off between the risk of dying from cancer versus remaining on lifelong dialysis treatment 6.

With a greater accessibility to the growing armamentarium of novel anti‐cancer therapies combined with improved overall survival of cancer patients in the general population, the current recommendations and clinical practice for determining transplant suitability of patients with prior treated advanced cancers may no longer be applicable. This review will focus on a few of the controversial aspects of transplanting patients with prior cancers, including those with ESKD attributed to multiple myelomas.

Epidemiology of cancer incidence and mortality after kidney transplantation

Several large population cohort studies have consistently shown that cancer incidence is up to 100‐fold higher in solid organ transplant recipients compared to age‐ and gender‐matched general population, particularly for oncogenic virus‐related cancers 7, 8. Even though all solid organ transplant recipients have an excess risk of a large number of cancers post‐transplant, there are organ‐specific differences where the absolute risks of certain types of cancers are higher in subgroups of organ transplant recipients (e.g. the risk of renal cell cancer is higher in patients who have received kidney transplants). Impairment of cell‐mediated immunity arising from chronic exposure to immunosuppression is a major contributing factor in the pathogenesis of certain cancers post‐transplant, particularly cancers attributed or suspected to be related to a viral aetiology such as Epstein–Barr virus (EBV) and human herpesvirus 8 (HHV‐8) 9, 10. In addition, there is an increased risk of other cancer subtypes such as melanoma, and lung and renal cell cancers, where uraemia, inflammation and/or ability of cancer cells to escape immune surveillance are likely to be important in the development of these cancers. In contrast, the risks of common cancers such as breast and prostate cancers are not increased in kidney transplant recipients.

The prognosis of kidney transplant recipients who have developed incident cancer is relatively poor, with the risk of cancer mortality substantially greater compared to age‐matched general population 11, 12, 13. In two large contemporaneous population cohorts of kidney transplant recipients from the United Kingdom (UK, 2001–2012) and Australia and New Zealand (1980–2014), standardized mortality ratios (SMRs) were substantially higher in kidney transplant recipients of either gender and across all age groups, with SMRs highest for younger recipients 12, 13. Relative to the general population, the greatest SMRs of cancer deaths were non‐Hodgkin lymphoma [10.7, 95% confidence interval (95% CI) 8.9–12.7], kidney cancers (7.8, 95% CI 5.9–10.0) and melanoma (5.8, 95% CI 4.5–7.3) for kidney transplant recipients in Australia and New Zealand 12, whereas in the UK, lung cancer and lymphoma were the most common cause of cancer deaths in men and women, respectively 13. There was no significant increased risk of cancer death from breast cancer in women or prostate cancer in men. Similar findings have been corroborated in population cohorts from Canada and Hong Kong 14, 15.

Cancer recurrence post‐kidney transplant in patients with prior cancers

The risk of cancer recurrence post‐transplant in patients with prior cancer remains poorly described. In a population cohort study of 2840 kidney transplant recipients from Australia and New Zealand who have developed cancer post‐transplant between 1965 and 2012, 80 (2.8%) had a history of prior cancers, with 23 (0.8%) and 57 (2.0%) recipients who experienced cancer recurrence or had developed a second new cancer, respectively 16. The most frequent recurring cancers after transplantation included urinary tract cancers (30%), followed by breast cancers (26%), melanoma (13%) and female genital tract cancers (13%). However, it appears that the cumulative cancer‐specific and patient survivals were similar between recipients who had developed de novo cancers and those who had experienced cancer recurrence, but these findings, however, were limited by the small event rates. In a French cohort study of 143 patients with prior renal cell cancer, cancer recurrence rate post‐kidney transplant was 9% (13 of 143 patients), with over 75% of patients who had experienced cancer recurrence died during the follow‐up period 17. In a systematic review of 57 studies (37 studies in the meta‐analyses) reporting on cancer recurrence in solid organ transplant recipients (19 studies in kidney transplant recipients), the proportion of patients (395 recurrences in 5838 patients) who had experienced recurrent cancer ranged between 0.4% and 22% (the latter from Israel Penn registry, a voluntary reporting registry) 18, 19, with the pooled proportion of patients with recurrence of 5.0% (95% CI 3.2–7.6%). Thirteen of the 57 studies were in abstract form and 17 of the remaining 44 studies were deemed to be of low quality. The pooled estimates for site‐specific recurrence rate for all solid organ transplant recipients were uncertain and imprecise. The highest recurrence rate was reported for lung cancer (5.4 events per 100 person‐year, 95% CI 1.7–16.6), gastrointestinal cancer (4.7 events per 100 person‐year, 95% CI 1.7–12.4), cervical cancer (3.9 events per 100 person‐year, 95% CI 1.6–9.3), kidney cancer (2.2 events per 100 person‐year, 95% CI 0.8–6.2), breast cancer (2.0 events per 100 person‐year, 95% CI 1.3–3.3) and melanoma (1.9 events per 100 person‐year, 95% CI 0.8–4.7). In studies confined to only kidney transplant recipients (maximum of four studies for each site‐specific cancer), the cancer recurrence rate per person‐year of follow‐up was highest for kidney cancer (seven recurrences per 334 person‐year of follow‐up), melanoma (7 per 235 person‐years), breast cancer (4 per 83 person‐years), cervical cancer (3 per 98 person‐years), gastrointestinal cancer (2 per 20 person‐years) and thyroid cancer (2 per 36 person‐years) 20. It is noteworthy that the pattern and frequency of cancer recurrence after kidney transplantation is dissimilar between studies, but is likely to reflect differing approaches in data reporting of cancer recurrence (i.e. reports of cumulative incidences, relative/absolute recurrence rates); dissimilar risk factors (e.g. analgesic nephropathy as cause of ESKD is more common in Australia and parts of Europe, which predispose to a greater risk of renal cell and bladder cancers) 21, 22, 23, availability and/or uptake of cancer screening programmes, differences in country‐specific patient characteristics/ethnicity, potential differences in the geographical distribution of oncogenic viruses; and the inclusion of other nonkidney solid organ transplant recipients in the studies. Nonetheless, it is likely that the reporting of cancers to country‐specific transplant registries is often incomplete, which may lead to an underestimation (and differences between countries) of the cancer recurrence risk.

Recommendation for transplant eligibility and ‘optimal’ waiting time prior to transplant wait‐listing

The current recommendation regarding medical suitability of patients with prior cancers and the duration of waiting time between the treatment of prior cancer(s) and time to wait‐listing/transplantation is primarily extrapolated (and arbitrarily established) from the data of potential cancer recurrence in the general population. Therefore, the benefit and cost of cancer recurrence relative to the projected patient and allograft survival compared to no transplantation are not explicitly considered and the risk of cancer recurrence is anticipated to be increased (but relatively unknown) in the setting of chronic immunosuppression. Given that the risk of cancer recurrence and recurrent cancer‐related mortality appears low from previous epidemiological studies 16, it is unclear whether the current recommendation is justified or whether we are too restrictive in determining the transplant eligibility of patients with prior cancers, and whether a lesser waiting time may be appropriate in certain cancers where the survivals have substantially improved in the general population. Nevertheless, clinicians and patients will need to be cognizant that treatment for cancer recurrence in kidney transplant recipients is exceedingly dissimilar to the general population, taking into consideration the potential drug interaction between chemotherapeutic agents with immunosuppressive regimen, modification of immunosuppressive regimen in the presence of chemotherapeutic agents, combined cumulative toxicities (including infectious complications) of the addition of chemotherapeutic agents to immunosuppressive regimen, suboptimal kidney allograft function necessitating reduction in the dose of chemotherapeutic agents (which may reduce the efficacy of treatment) and the potential to induce kidney allograft rejection and premature allograft failure as a result of under‐immunosuppression (reducing or withdrawal of specific immunosuppressive agent or as a result of inducing acute rejection with the use of immune‐stimulatory therapy) 24. Not infrequently, clinicians and patients must decide on the ‘paradoxical’ trade‐off between sustaining allograft function and the adequate treatment of cancer recurrence at the expense of a functioning allograft.

The recommended waiting time for candidates with prior solid organ cancers varied between 2 and 5 years, depending on the cancer types and stage. Table 1 shows a number of guidelines highlighting the transplant eligibility of patients with selected prior cancers where there is a notable disparity in the recommendations. To assess the quality of each individual guideline, the Appraisal of Guidelines for Research & Evaluation (AGREE) II instrument should be used. The various domains of the AGREE II include the scope and purpose, stakeholder involvement, rigour of development, clarity of presentation, applicability and editorial independence 25. Each item was rated on a 7‐point scale ranging from strongly disagree (score 1) to strongly agree (score 7). The domain score was obtained by summing all scores of the individual items per domain and then standardizing the total as a percentage of the maximum possible score for that domain. The American Society of Transplantation (AST) and European Renal Best Practice (ERBP) guidelines (specifically evaluating the section relating to prior cancers) would have rated relatively well across the six domains, whereas for the Caring for Australasians with Renal Impairment (CARI) guidelines, the quality of the domains was considered inadequate and under‐developed. Nevertheless, none of the guidelines were able to assess in detail all types and stages of cancers, which may reflect the lack of supporting data for many of these cancers.

Table 1.

Eligibility guidelines for kidney transplantation for patients with treated prior early and advanced cancers.

Cancer type/stage	Guidelines
	AST 2001 27	CST 2005 28	ERBP 2013 29	CARI 2013 30
Breast cancer
Stage 0–2 (early)	√(2)	√(2–5)	?	√(2)
Stage 3–4 (advanced)	√(5)	Χ	?	Χ
Colorectal cancer
Duke C	√(2–5)	√(5)	??	√(5)
Duke D	?	?		Χ
Melanoma
In situ	√(2)	√(2)	?	√(2)
Invasive	√(5)	√(5)	√(5)	√(5)
Renal cell cancer
Early/symptomatic	√(0–2)*	√(0–2)	√(1)	√(5)
Large/invasive	√(5)*	√(5)	?	?
Multiple myeloma	?	Χ	?	Χ
Lymphoma	√(2)	√(2)	√(1–3)	√(2)
Lung cancer
Localized	√(2)	√(2)	√(5)	?
Invasive	?	?	?	?
Thyroid cancer†	√(2)	√(2)	√(1–3)	√(2)
Bladder cancer
Invasive	√(2)	√(2)	?	√(2)
Nonmelanoma skin cancers
Localized	√(0)	√(0)	√(0)	√(0)
Invasive	?	?	?	?

CST, Canadian Society of Transplantation; ERBP, European Renal Best Practice; √(0) denotes no waiting time, √(2) denotes waiting time of at least 2 years, √(5) denotes waiting time of at least 5 years, √(2–5) denotes waiting time of at least 2–5 years, Χ denotes contraindication for transplantation, ? denotes no recommendation or unknown; *Symptomatic renal cell cancer ≥2 years after surgical removal, large (≥5 cm) and/or invasive renal cancers 5 years of waiting time, no waiting time required if <5 cm discovered incidentally, †Tumour invasiveness not specified.

The 2018 Kidney Disease Improving Global Outcomes (KDIGO) clinical practice guideline on the evaluation and management of candidates for kidney transplantation is currently available for public review and is likely to incorporate updated guidelines regarding suitability of transplant wait‐listing for patients with prior cancers. In contrast to the reported updated recurrence rate of cancers in the general population 26, the recurrence rates for patients with prior cancers remain poorly defined and with data primarily from the Israel Penn registry in the 1990s, which therefore does not take into account the cancer histology, updated staging and classification, evolution of improved treatments and survival of patients with these cancers 18, 27. Table 2 shows the cancer stage‐specific survival, recurrence rate and prognostic characteristics in the general population ± kidney transplant recipients, along with a proposed recommendation based on the available data. Given the uncertainties, the recommended transplant suitability and waiting time must be individualized, informed by a multidisciplinary specialty team‐centred approach involving the haematologists and oncologists (where appropriate) and engagement of the patients, based not solely on cancer staging but also on prognostic characteristics of the tumour (genetic, clinical and biochemical/haematological), expected survival on dialysis versus transplantation, projected risk of disease recurrence and the beliefs and expectations of both clinicians and patients regarding the benefits of kidney transplantation.

Table 2.

Survival, recurrence and prognostic features of site‐specific cancer types in the general population and kidney transplant recipients

Cancer type	SEER* (5‐year survival)	AJCC (5‐year survival)	Recurrence rate (general population)	Recurrence rate (kidney transplant)	Prognostic features	Recommendation
Breast cancer 31, 32	Localized – 99% Regional – 85% Distant – 27%	Overall survival: Stage I – 94–97% Stage IIA – 88–97% Stage IIB – 91–100% Stage IIIA – 67–100% Stage IIIC – 33–84%	10‐year recurrence rate: T1N0 stage – LR rates 4.6%, RR 2.3% and DM 7.8%. T2N1 stage – LR rates 6.2%, RR 5.2% and DM 19.6%. Recurrence risk reduces with increasing time.	23%	Hormone receptor status, HER‐2 receptor status, gene expression (Oncotype DX® and the MammaPrint®), histological grade and ploidy and cell proliferation	Stage I/II Stage III Stage IV
Colorectal cancer 33, 34	Localized – 90% Regional – 71% Distant – 14%	Overall survival: Stage I – 83% Stage II – 70% Stage III – 58%	5‐year recurrence rate: Stage I – 5% Stage II – 12% Stage III – 33%	21% (mortality rate from recurrence 63%)	Tumour location, grade, size, number of LNs involvement	Stage I/II Stage III Stage IV
Melanoma 35, 36	Localized – 98% Regional – 64% Distant – 23%	In situ – 99–100% Stage I – 97–99% Stage II – 82–94% Stage III – 32–93%	Crude 5‐year recurrence/progression‐free survivals: Stage I – 85% Stage II – 59% Stage III – 17% Stage IV – 13%	21%	Tumour thickness, presence of ulceration, mitotic index, extent of nodal metastases, extent of satellitosis and in‐transit disease, presence of tumour ulceration, LDH level	Stage I Stage II/III Stage IV
Renal cell cancer 17, 37, 38	Localized – 93% Regional – 69% Distant – 12%	Overall survival: Stage I – 81% Stage II – 74% Stage III – 53% Stage IV – 8%	Recurrence rate (80% recurs within 3 years): T1 tumour – 7% T2 – 26% T3 – 39%	27% (1% if incidental cancer; cumulative incidence of recurrence: 8% at 5 years and 15% at 10 years)	Age, LDH level, serum calcium level, haematological parameters, tumour biology and stage, histological clear cell type, extent of ‘m’ = nodal and distant metastases	Stage I/II Stage III Stage IV
Lung cancer 39, 40, 41, 42, 43	SCLC Localized – 29% Regional – 15% Distant – 3% NSCLC Localized – 60% Regional – 33% Distant – 6%	SCLC Stage I – 46–54% Stage II – 34–36% Stage III – 8–26% Stage IV – 7% NSCLC Stage I – 53–68% Stage II – 36–41% Stage III – 7–20% Stage IV – 3%	NSCLC: 30% to 55% of patients develop recurrence Population cohort (all lung cancer subtypes) – recurrence rates: Stage IA to IB – about one‐third of patients had a recurrence Stage IIA – 61.2% Stage IIB – 57.9% Stage IIIA – 62.8%	Unknown	Age, nodal involvement, performance status, weight loss, tumour size, LDH level, sodium level,	Stage I Stage II–IV
Lymphoma 44, 45, 46	NHL Localized – 72% Regional – 72% Distant –55% HD Localized – 92% Regional – 93% Distant –78%	NHL Risk categories: Low – 73% Low‐intermediate – 51% High‐intermediate – 43% High – 26% All subtypes: Stage I – 70% Stage II – 72% Stage III – 58% Stage IV – 57% HD Stage I – >90% Stage II – >90% Stage III – ≥80% Stage IV – ≥65%	Recurrence variable (up to 50% for aggressive lymphoma)	11%	Age, cell type, tumour bulk, extranodal involvement, LDH level, performance status, presence of B symptoms and response to initial treatment	All stages ‘Aggressive’
Thyroid cancer 47, 48, 49	Papillary Localized – 100% Regional – 100% Distant – 78% Follicular Localized – 100% Regional – 96% Distant – 56% Anaplastic Localized – 30% Regional – 13% Distant – 3%	10 years disease‐specific survival: Stage I – 98–100% Stage II – 85–95% Stage III – 60–70% Stage IV – <50%	Up to 20% (for papillary and follicular cancers)	7%	Age, tumour size, ≥2 lymph nodes involvement, bilateral tumours	Stage I/II Stage III Stage IV or anaplastic
Bladder cancer 50, 51, 52	Localized – 69% Regional – 35% Distant – 5%	Stage I – 85–90% Stage II – 30–45% Stage III – 15–30% Stage IV – 10%	Overall 2‐year, 5‐year and 10‐year recurrence rates: 61%, 69% and 74%, respectively	29%	Age, female gender (higher risk), Black race, undifferentiated or higher grade tumour	Stage I Stage II–IV
NMSC (SCC) 53, 54, 55	Not reported	No nodal involvement – 96% Nodal involvement + adequate treatment – 72% Nodal involvement + no treatment – 25–35%	Variable, with recurrence rate of 7% ay 5 years (majority within 2 years). ‘Higher risk’ tumour higher recurrence rate (perineural invasion 19%)	48%	Size, cancer borders, primary/recurrent cancer, presence of immunosuppression, pathology (tumour differentiation; thickness/depth; perineural, vascular or lymphatic involvements), location, size, growth pattern	Localized – none Invasive Nodal/distant metastasis

AJCC, American Joint Committee on Cancer; DM, distant metastasis; HD, Hodgkin disease; HER‐2, human epidermal growth factor receptor‐2; LDH, lactate dehydrogenase; LR, local recurrence; NHL, non‐Hodgkin lymphoma; NMSC, nonmelanoma skin cancer; NSCLC, non‐small‐cell lung cancer; RR, regional recurrence; SCC, squamous cell cancer (skin); SCLC, small cell lung cancer; SEER, The Surveillance, Epidemiology and End Results. Waiting time of at least 1 year; Waiting time of at least 5 years; Contraindication for transplantation unless exceptional clinical circumstances. *Survival data extracted from American Society of Cancer website (ref. 22).

By way of an example, the guidelines consider the staging of prior breast cancer as a simplistic classification of localized versus advanced and the recommendation applicable to the stage of the cancer when it is first diagnosed. In addition, the standardized tumour, node and metastases (TNM) classification system, which considers in more details the size, nodal involvement and spread of the tumours (for breast cancer: stages IA, IB, IIA/B, IIIA–C and IV), provides a more accurate description and prognosis of the cancers. Other factors such as age, performance status, response to treatment and presence of certain tumour characteristics [such as hormone receptor and human epidermal growth factor receptor‐2 (Her2) status for breast cancer] will influence prognosis, further compounding the complex decision when determining whether patients with prior cancers are suitable for transplantation (Table 2). The current guideline recommendations are often based on low or ungraded quality of clinical evidence extrapolated from cohort studies and/or the opinions of the expert committee, and therefore, clinicians, in collaboration with haematologists and oncologists, should consider these guidelines as a supportive adjunct in the clinical decision‐making when considering the eligibility of patients with prior cancers for transplantation.

Screening and surveillance for patients with prior cancers

In a critical appraisal of the clinical cancer screening guidelines for kidney transplant recipients, Wong et al. 56 indicated that the current recommendations for cancer screening in kidney transplant recipients were exclusively extrapolated from study findings in the general population (with no randomized controlled trials undertaken in the kidney transplant recipients), and as such, there were insufficient data to inform cancer screening guidelines in kidney transplant recipients pre‐ and post‐transplant. Complex simulation model has shown a modest benefit of cancer screening in kidney transplant recipients, with the incremental cost‐effectiveness ratio of cancer screening for colorectal, breast and prostate cancers in an average risk nondiabetic White population between $25 000 and $57 000USD per life‐years saved 57. There may be incremental benefits in higher risk transplant recipients, but given the assumptions and limitations of the model as well as the imprecise estimation of life expectancy of these patients, the evidence to support cancer screening in this population remains weak. Similar magnitude of benefits and uncertainties have been shown in a model of implementing routine ultrasound screening to detect asymptomatic renal cell cancer in kidney transplant recipients 58. One recent prospective cohort study showed that faecal occult blood testing (FOBT) was an accurate screening test to detect advanced colorectal neoplasia in kidney transplant recipients; with test sensitivity, specificity, positive predictive value and negative predictive value of 97%, 82%, 26% and 98%, respectively 59. Nevertheless, there are continuing uncertainties relating to the diagnostic test performance of other population‐based cancer screening techniques including cervical cancer screening test and mammography; the competing risks of cardiovascular and infection‐related mortality; the projected reduced survival of patients with ESKD; and the beliefs and concerns of patients regarding cancer screening relative to their health needs and priorities, and therefore, the strengths of the current recommendations for cancer screening in the kidney transplant recipients remain indeterminate 56, 60. Consequently, the timing, optimal imaging technique and frequency of targeted cancer screening to verify persistent cancer clearance prior to and during wait‐listing, or the timing and frequency of targeted cancer screening after kidney transplantation remain unknown, although most clinicians would recommend for more frequent targeted screening in these patients in the absence of definitive evidence or guidelines. In a systematic review comprising of 16 cohort studies (inclusive of all solid organ transplant recipients, with eight cohort studies in kidney or kidney–liver and kidney–pancreas transplant recipients), patients with prior cancer were found to have up to a twofold greater risk of developing de novo cancers post‐transplantation [hazard ratio (HR) 1.92; 95% CI 1.52–2.42] 61. Given the current available evidence, adherence to age‐specific population‐based screening should be encouraged, but there are insufficient data to suggest that more frequent targeted or general cancer screening will be cost‐effective or of clinical benefit in kidney transplant recipients with prior cancers.

Management approach of patients with prior cancers post‐kidney transplantation

The direct and indirect carcinogenic effects of the cumulative burden of immunosuppressive agents are well established, which involves a complex interplay of an altered immune balance favouring a reduction in tumour immune surveillance and the activation of human oncogenic‐type viruses 62. Consequently, clinicians often advocate a lower cumulative burden of immunosuppression for patients with prior cancers, either by allocating donor kidneys associated with a lower immunological risk [e.g. better human leucocyte antigen (HLA)‐matched kidneys or the absence of pretransplant donor‐specific anti‐HLA antibody], thereby allowing the use of reduced‐intensity immunosuppressive regimen, or by avoiding the excess utilization of potent immunosuppressive agents such as T‐cell‐depleting antibody which has shown to be associated with a heightened risk of incident cancer post‐transplant 63, 64. However, the critical balance between maintaining an adequate amount of ‘lower burden’ immunosuppressive regimen versus the risk of acute and chronic rejection is often difficult to achieve or predict; and frequent immunological risk assessment post‐transplant such as the monitoring of de novo donor‐specific anti‐HLA antibody or protocol biopsy to detect subclinical rejection may assist in the clinical decision‐making process of individualizing immunosuppression for these patients.

Mammalian target of rapamycin (mTOR) inhibitor has been shown to possess antiviral and anti‐tumour properties, with epidemiological and trial data showing that maintenance treatment with mTOR inhibitor in kidney transplant recipients significantly reduced the risk of skin and possibly nonskin cancers 65, 66, 67. Several systematic reviews of randomized controlled trials have shown that the early conversion from CNI treatment to mTOR inhibitor (±low‐dose CNI) reduced the risk of incident cancer post‐kidney transplant, compared to CNI treatment regimen, but this risk reduction was observed predominantly for nonmelanoma skin cancers (NMSC) 68, 69. Several randomized controlled trials have also shown that mTOR inhibitor was effective in the secondary prevention of NMSC in kidney transplant recipients, but the long‐term tolerability of mTOR inhibitor remains relatively poor 70, 71, 72. In addition, several studies have shown that switching from CNI treatment to mTOR inhibitor may be effective in tumour regression in patients who have developed post‐transplant lymphoproliferative disease (PTLD), but this approach has yet to be tested in a randomized controlled trial 73, 74. Given the continuing uncertainty regarding the efficacy of mTOR inhibitor in reducing the risk of nonskin cancers, the preferential use of mTOR inhibitor immunosuppressive regimen cannot be recommended for all patients with prior cancers; with the clinical decision based on the expected benefit as opposed to the potential long‐term adverse effect of mTOR inhibitor including a higher risk of allograft failure, mortality, development of de novo donor‐specific anti‐HLA antibody and a higher risk of antibody‐mediated rejection compared to CNI treatment 75, 76, 77, 78, 79.

In patients who have developed de novo cancers during prior kidney transplants, the clinical decision regarding medical suitability and duration of cancer waiting time when considering repeat transplantation should be similar to the current recommendations of those with a history of cancers prior to the initial transplantation. For certain transplant‐specific cancers such as Kaposi's sarcoma and PTLD, successful repeat transplantation has been consistently reported although reports of cancer recurrence (for both Kaposi's sarcoma and PTLD) have been infrequently reported 80, 81, 82, 83, 84, 85, 86. However, the timing of repeat transplantation, the optimal immunosuppressive regimen (or whether to consider using mTOR inhibitor as initial immunosuppression) or whether certain prior cancer‐specific characteristics (e.g. Epstein‐Barr virus [EBV] positivity, initial response to treatment) influence the re‐transplant outcomes of patients with prior Kaposi's sarcoma and PTLD remains poorly defined.

‘High‐risk’ pretransplant cancers: multiple myeloma and nonlocalized melanoma

The current guidelines suggest that patients with multiple myeloma or with nonlocalized melanoma should not be considered for kidney transplantation, with the assumption that the survival of these patients will be severely reduced post‐transplant as a result of cancer recurrence. In the presence of improved survival of patients with multiple myeloma and advanced malignant melanoma, combined with the availability of novel anti‐neoplastic agents and increasing number of reports of acceptable short‐ and medium‐term outcomes post‐transplant in patients with these cancers, there are increasing challenges to the long‐standing paradigm of excluding patients with multiple myeloma and nonlocalized malignant melanoma who have achieved complete cancer remission for kidney transplantation.

Kidney transplantation in patients with multiple myeloma

With the evolution and availability of more potent anti‐neoplastic agents combined with improved survival of patients with multiple myeloma 87, 88, kidney transplantation is now a practical and conceivably an underutilized treatment option. Prior case reports describing the outcome of patients treated with chemotherapy (e.g. dexamethasone, bortezomib) to achieve clinical remission prior to kidney transplantation have largely been disappointing, with high rates of disease relapses (up to 70%) requiring further treatment. In patients who had experienced recurrence, the time to relapse post‐transplant varied between 3 months and over 3 years, with a proportion dying from disease recurrence or treatment‐related infective complications 89, 90, 91, 92, 93, 94. The change in treatment strategy to include preconditioning chemotherapy followed by high‐dose alkylating agent and autologous stem cell transplant (ASCT) to achieve complete remission prior to kidney transplantation has resulted in favourable short‐ to intermediate‐term outcomes and low rates of relapse, suggesting that treatment strategy to include ASCT may be necessary to achieve sustained clinical remission post‐transplant (n = 23 cases; Table 3). Nonetheless, there are several questions which remain unanswered: (i) the ‘optimal’ time of clinical remission post‐ASCT prior to kidney transplantation, (ii) clinical or haematological risk factors for disease relapse after kidney transplant, (iii) consideration of the timing for kidney transplantation and differences in post‐transplant outcomes between those who have achieved complete versus very good partial response, (iv) clinical benefit in maintenance anti‐myeloma treatment post‐kidney transplant and (v) kidney transplant eligibility of patients with disease affecting extrarenal organs (e.g. heart). It is likely that the answers to these questions may be partly addressed as more cases of patients with multiple myeloma are considered for kidney transplantation and longer‐term outcomes are ascertained.

Table 3.

Kidney transplant outcomes of patients with multiple myeloma in the era of autologous stem cell transplantation

	Number of cases (follow‐up time)	Time from diagnosis to kidney transplantation (pretransplant treatment)	Relapse post‐transplant (time from kidney transplant to relapse and treatment)	Allograft and patient outcomes
Shah et al. 95	N = 5 ISS III (median follow‐up 55 months)	Median 27 months (high‐dose melphalan and ASCT)	2/5 relapsed (median 11 months) and died (median of 52 months) post‐transplant despite chemotherapy	80% at 4 years
Batalini et al. 96	N = 4 with MIDD (unknown, 2/4 CR at time of transplant)	Median 2.6 years after treatment (high‐dose melphalan and ASCT)	Not reported	Not reported
Le et al. 97	N = 4 LD transplants, 2/4 cases LCDD kidney biopsy (f/up range 16–58 months)	Range 20–66 months (high‐dose melphalan and ASCT)	1/4 stable disease (not CR prior to transplant), bortezomib → lenalidomide maintenance	Survivals 100% at f/up
Hassoun et al. 98	N = 2 LCDD (f/up range 26–70 months)	Treatment to transplant 14–46 months (high‐dose melphalan and ASCT)	None	Survivals 100% at f/up
Sanchez‐Quintana et al. 99	N = 2 (1 LCDD, 1 MM; f/up range 3–4 years)	Treatment to transplant 4 years (ASCT with CR or VGPR pretransplant)	None, both received lenalidomide maintenance	Survivals 100% at f/up
Girnius et al. 100	N = 2 (1 MIDD; f/up 1 and 5 years)	Not reported (high‐dose melphalan and ASCT)	None	Survivals 100% at f/up
Khoriaty et al. 101	N = 1 MM (f/up 1 year)	3.5 years (high‐dose melphalan and ASCT)	None	Survival 100% at f/up
Royer et al. 102	N = 1 LCDD (f/up 50 months)	36 months (high‐dose melphalan and ASCT)	None	Survival 100% at f/up
Lorenz et al. 103	N = 1 LCDD (f/up 23 months)	9 months (high‐dose melphalan and ASCT)	None	Survival 100% at f/up
Bansal et al. 104	N = 1 MIDD (f/up 2 years)	<1 year (high‐dose melphalan and ASCT)	None	Survival 100% at f/up

ASCT, autologous stem cell transplant; CR, complete response; f/up, follow‐up; ISS, Multiple Myeloma International Staging System; LCDD, light‐chain deposition disease; LD, live donor; MM, multiple myeloma; MIDD, monoclonal immunoglobulin deposition disease; VGPR, very good partial response.

Kidney transplant eligibility of patients with advanced malignant melanoma

Malignant melanoma is the most fatal type of skin cancer affecting the general population, with global incidence steadily increasing over the last decade 105, 106. Tumour characteristics such as cancer stage, presence of ulceration and thickness of the tumour have prognostic significance, and patients with more invasive disease have a higher risk of cancer recurrence and melanoma‐specific mortality 107. In patients who have received solid organ transplants, the risk of developing de novo melanoma is twice that of age‐ and gender‐matched general population, likely resulting from the chronic suppression of cell‐mediated immunity attributed to lifelong immunosuppressive therapy 7, 108, 109. There is consistent epidemiological evidence to suggest that kidney transplant recipients with melanoma have significantly higher Clark's level and pathological staging, with up to four times the risk of cancer mortality compared to the general population (Table 4) 12, 110. Consequently, clinicians are often reluctant to consider kidney transplantation in patients with prior history of melanoma, with those with invasive melanoma [American Joint Committee on Cancer (AJCC) stage II and above] considered absolute contraindication for transplantation. Given the improved survival and treatment options for patients with melanoma in the general community 111, there is considerable debate regarding the acceptability of patients with more advanced treated melanoma for transplantation. In the general population, 40% of patients with treated melanoma and without metastatic disease will experience cancer recurrence, typically within the first 12 (for local and nodal recurrences) to 24 (for distant metastasis) months 112.

Table 4.

Survival rates of patients with malignant melanoma in the general population and after kidney transplantation

	General population 113, 114 (5‐year survival rate)*	Survival (kidney transplant recipients with de novo melanoma versus matched cohort from general population) 110 5‐year survival from melanoma diagnosis
		Kidney transplant	Matched cohort
AJCC staging
In situ	99–100%	–	–
Stage I		78% (stage I)†	98% (stage I)†
Stage IA (<0.8 mm)	99%	Adjusted HR melanoma‐specific death 3.55 (95% CI 1.09, 11.54)	Referent
Stage IB (0.8–2 mm)	97%
Stage II		35% (stage II)†	62% (stage II)†
Stage IIA (2.01–4 mm)	94%	Adjusted HR melanoma‐specific death 1.30 (95% CI 0.49, 3.45)	Referent
Stage IIB (>4 mm)	87%
Stage IIC (>4 mm + ulceration)	82%
Stage III		20% (stage III/IV)†	32% (stage III/IV)†
Stage IIIA	93%		Referent
Stage IIIB	83%	Adjusted HR melanoma‐specific death 1.21 (95% CI 0.27, 5.41)
Stage IIIC	69%
Stage IIID	32%
Stage IV	15–20%

95% CI, 95% confidence interval; AJCC, American Joint Committee on Cancer; HR, hazard ratio; mm, millimetres. *Survival rate from time of diagnosis in people with malignant melanoma between 2008 and 2014. †Approximate estimates only.

Table 5 shows the current evidence summarizing the risk of cancer recurrence after kidney transplantation in those with a prior history of melanoma. The recurrence risk in those with prior melanoma varies between 0% and 35%, but intermediate‐ to long‐term allograft and patient survivals are poorly defined. It is noteworthy that details relating to the pathological staging of melanoma pretransplant, treatment and the timing between melanoma diagnosis/treatment and transplantation are inadequately described, with the majority of the cohorts having included patients prior to 2000. The current recommendations regarding the suitability and waiting time for patients with prior melanoma remain inconsistent and largely extrapolated from population cohort studies from earlier eras, with an over‐simplified broad classification staging system and other prognostic factors (Table 1). With the available data, it is difficult to ascertain whether the current recommendations are appropriate (given the low rates of melanoma recurrence post‐transplant in current clinical practice) or whether the guidelines are too restrictive in excluding those patients with higher staged melanomas who may still derive a relative survival benefit from transplantation compared to dialysis treatment or death from melanoma recurrence.

Table 5.

Incidence of recurrent melanoma after kidney transplantation in patients with prior history of melanoma

Author (year)	Study type	Recurrence (time)	Outcome
Dahle 115 (1963–2010)	Single centre Norway (kidney only)	3 of 20 (unknown)	Similar survival prior cancers (all cancers) versus no prior cancers, higher risk of cancer deaths
Hellstrom 116 (1982–2013)	Single centre Sweden (kidney only)	1 of 6 (unknown)	No cancer death reported
Viecelli 16 (1965–2012)	ANZDATA Registry (kidney only)	3 of 8 (unknown)	1 melanoma recurrence death
Unterrainer 117 (1984–2016)	CTS registry (kidney only)	8 of 164 (unknown)	Similar graft and patient survivals prior melanoma versus no prior cancers, higher risk of cancer deaths with HR of 2.56 (95% CI 1.51–4.34) up to 10 years
Chapman 118 (1963–1999)	ANZDATA Registry (kidney only)	2 of 19	Not reported
Brewer 119 (1967–2007)	Mayo Clinic databases, OPTN, IPITTR (kidney only)	2 of 59	2× recurrences (1 nodal and 1 lung)
Matin 120 (1976–2007)	Europe SCOPE network (kidney only)	0 of 9	Follow‐up 14 years postmelanoma. Interval from melanoma to transplant range: 0.4–33 years. Breslow thickness range: 0.5–18 mm (3 no records)
Puza 121 (2001–2016)	Duke University, US. All solid organs)	2 of 12 (median time 5.3 years). 1 in situ and 1 stage IIIA	None with recurrence died from cancer. Median time between melanoma diagnosis and transplantation was 4.13 years (range: 1.1–13.3 years).
Acuna 122 (1991–2010)	CORR registry (all solid organs)	Not reported	All‐cause mortality: pretransplant melanoma (HR 1.76; 95% CI 1.12–2.77) versus matched cohort without pretransplant cancer
Arron S 123 (1987–2010)	SRTR database (all solid organs)	336 (112 in situ and 224 invasive)	Pretransplant melanoma: death because of melanoma (adjusted HR 27, 95% CI 11–64; P < 0.0001), all‐cause mortality (adjusted HR 1.3, 95% CI 1.0–1.5; P = 0.02); incident melanoma after transplant (adjusted HR 5.4, 95% CI 2.9–9.8; P < 0.0001)
Kang 124 (2005–2013)	UNOS database (kidney only)	398 with pretransplant melanoma	Pretransplant melanoma: post‐transplant cancer (any) (adjusted SHR 1.77, 95% CI 1.30–2.40; P < 0.001); post‐transplant skin cancer (adjusted SHR 1.93, 95% CI 1.38–2.69; P < 0.01)
Dapprich 125 (1978–2007)	Mayo clinic database (all solid organs)	0/12 recurrence. Pretransplant AJCC: stage 0 (n = 4), IA (n = 3), IB (n = 2), IV (n = 1); unknown (n = 2)	No deaths from melanoma/cancer. Median time from melanoma to transplantation 3.8 years
Penn 126 (1968–1995)	IPITTR (all solid organs)	6/31 recurrence (n = 3 within 2 years post‐transplant, n = 2 between 2 and 5 years)	Not reported

95% CI, 95% confidence interval; AJCC, American Joint Committee on Cancer; ANZDATA Registry, Australia and New Zealand Dialysis and Transplant registry; CORR, Canadian Organ Replacement Register; CTS, Collaborative Transplant Study; HR, hazard ratio; IPITTR, Israel Penn International Transplant Tumor Registry; mm, millimetres; OPTN, Organ Procurement and Transplantation Network; SCOPE, Skin Care in Organ Transplant Patients, Europe; SHR, subdistribution hazard ratio; SRTR, Scientific Registry of Transplant Recipients; UNOS, United Network for Organ Sharing.

Implications for future research

In the absence of adequate clinical evidence to support the recommendations of transplant eligibility or waiting time across all cancer types, the creation of a global repository of patients with all cancer types (including accurate records of the updated histology, prognostic factors, staging and classification and response to treatment) for patients with EKSD (dialysis and transplantation) will help inform future clinical guidelines. There continues to be uncertainty as to the applicability of age‐ and gender‐specific general population cancer screening guidelines to patients with ESKD pre‐ and post‐transplantation and future research examining the test performance of cancer screening in this population should be prioritized 59, including the appropriateness and cost‐effectiveness of screening renal tract ultrasound (to detect renal cell carcinoma) and lung imaging (to detect lung cancer particularly in higher risk population such as former/current smokers) given the higher incidence of these cancers in patients with ESKD 58. In addition, a greater emphasis on attaining effective consumer engagement along with a transparent public consultative process in clinical practice guideline development is critical, particularly on the standpoints of eligibility and ‘acceptable’ waiting times for patients with prior cancers (for both potential live and deceased donor kidney transplantation). Nevertheless, the integration of quality and diverse consumer participation in clinical guideline development or in healthcare research remains in its infancy, and the conception of a framework to establish optimal consumer engagement to inform evidence‐based healthcare guidelines and research is needed 127, 128.

Conclusion

The continuing dilemma facing organ transplant programmes is the imbalance between donor supply and demand, and therefore, each transplant programme will continue to prioritize and accept for transplantation only patients who are deemed clinically suitable to receive an organ. Patient preferences for how organs should be allocated is often not aligned with that of the community or clinicians, and often, the refusal to accept patients for transplantation who are otherwise suitable with the exception of a history of treated advanced cancer is often centred on the concern that cancer recurrence is likely to be at an increased risk and will lead to premature mortality and substantial reduction in the utility of the allograft, although the basis of this recommendation originates from outdated and poor‐quality data. This decision ignores the rights and ability of the patients to provide informed consent for treatment, balancing the projected survival benefit of transplantation (including improvement in quality of life) versus the likelihood of cancer recurrence, which is difficult to quantify accurately.

There is generally a lack of comparative data of the outcomes of patients with prior cancers if the waiting time after cancer treatment to transplantation was reduced, but evidence to support or refute such an approach is lacking and unlikely to be forthcoming. For patients without potential live kidney donors, a large proportion of patients on the deceased donor transplant waiting list will be required to wait on average of 2–5 years before transplantation for nonlocalized cancers, and clinical and transplant allocation programmes should consider factoring in the projected waiting time (for a kidney transplant) and patient survival when determining the ‘most appropriate’ waiting time prior to wait‐listing for patients with prior cancers. The decision to consider transplantation for those with prior treated advanced cancers should be individualized with wider consultation with the broader multidisciplinary team including haematologist or oncologist, clearly outlining the projected risk associated with cancer recurrence and the potential treatment options and outcomes if cancer does recur post‐transplantation.

Funding

The authors have declared no funding.

Conflicts of interest

The authors have declared no conflicts of interest.