Relative carcinogenicity of tacrolimus vs mycophenolate after solid organ transplantation and its implications for liver transplant care

Abstract

BACKGROUND

De novo malignancy is a leading cause of late morbidity and mortality in liver transplant recipients. Cumulative immunosuppression has been shown to contribute to post-transplant malignancy (PTM) risk. There is emerging evidence on the differential carcinogenic risk profile of individual immunosuppressive drugs, independent of the net effect of immunosuppression. Calcineurin inhibitors such as tacrolimus may promote tumourigenesis, whereas mycophenolic acid (MPA), the active metabolite of mycophenolate mofetil, may limit tumour progression. Liver transplantation (LT) is relatively unique among solid organ transplantation in that immunosuppression monotherapy with either tacrolimus or MPA is often achievable, which makes careful consideration of the risk-benefit profile of these immunosuppression agents particularly relevant for this cohort. However, there is limited clinical data on this subject in both LT and other solid organ transplant recipients.

AIM

To investigate the relative carcinogenicity of tacrolimus and MPA in solid organ transplantation.

METHODS

A literature search was conducted using MEDLINE and Embase databases using the key terms “solid organ transplantation”, “tacrolimus”, “mycophenolic acid”, and “carcinogenicity”, in order to identify relevant articles published in English between 1^st January 2002 to 11^th August 2022. Related terms, synonyms and explosion of MeSH terms, Boolean operators and truncations were also utilised in the search. Reference lists of retrieved articles were also reviewed to identify any additional articles. Excluding duplicates, abstracts from 1230 records were screened by a single reviewer, whereby 31 records were reviewed in detail. Full-text articles were assessed for eligibility based on pre-specified inclusion and exclusion criteria.

RESULTS

A total of 6 studies were included in this review. All studies were large population registries or cohort studies, which varied in transplant era, type of organ transplanted and immunosuppression protocol used. Overall, there was no clear difference demonstrated between tacrolimus and MPA in de novo PTM risk following solid organ transplantation. Furthermore, no study provided a direct comparison of carcinogenic risk between tacrolimus and MPA monotherapy in solid organ transplantation recipients.

CONCLUSION

The contrasting carcinogenic risk profiles of tacrolimus and MPA demonstrated in previous experimental studies, and its application in solid organ transplantation, is yet to be confirmed in clinical studies. Thus, the optimal choice of immunosuppression drug to use as maintenance monotherapy in LT recipients is not supported by a strong evidence base and remains unclear.

Key Words: Immunosuppression, Solid organ transplantation, Liver transplantation, Carcinogenicity, Tacrolimus, Mycophenolate

Core Tip: Cumulative immunosuppression exposure is an important risk factor for the development of post-transplant malignancy. There is emerging evidence on the differential carcinogenic risk profile of individual immunosuppressive drugs, independent of the net immunosuppression effect. This review demonstrates that the evidence on the relative carcinogenicity of tacrolimus and mycophenolic acid, the two agents most commonly used as maintenance monotherapy in liver transplant patients, remains unclear. Further studies are required to determine the clinical relevance of previous experimental findings to enable physicians to tailor immunosuppression regimens to minimize individual malignancy risk in solid organ transplantation.

INTRODUCTION

Liver transplantation (LT) remains the only curative treatment for end-stage liver disease and some cases of hepatocellular carcinoma, with an overall median survival of 20 years[1]. Despite improvements in short-term survival with the decline in rates of rejection and graft failure with the advent of modern immunosuppression regimens, long-term complications including post-transplant malignancy (PTM), have risen. Liver transplant recipients incur a 2- to 3-fold increase in rates of de novo malignancy compared to the general population[2,3]. Indeed, PTM has become a leading cause of late mortality in LT recipients[4,5].

The cumulative exposure to immunosuppression and direct carcinogenicity of individual agents may contribute to the development of PTM[6]. Tacrolimus and mycophenolic acid (MPA) are the most commonly used backbone immunosuppressants post-LT, and are also utilised as maintenance monotherapy in 42% of LT recipients in the United States due to the relatively immune tolerant microenvironment of the liver[7,8]. Experimental data have demonstrated multiple pro-oncogenic effects of tacrolimus, whereas MPA may be protective against tumour growth and progression[9]. This systematic review aims to compare the relative carcinogenicity of tacrolimus and MPA in solid organ transplantation to assist clinicians in making informed decisions regarding choice of immunosuppression regimens for patients.

PTM immunology

The development of PTM is a consequence of complex interactions between genetic, lifestyle and transplant factors (Figure 1). The central role of the immune system in cancer surveillance is highlighted by the increased malignancy risk that results from congenital and acquired immunodeficiencies, as well as the efficacy of immunotherapy for a growing number of malignancies such as hepatocellular carcinoma, melanoma, and renal cell carcinoma[6].

Figure 1 Risk factors for post-transplant malignancy.

An intact immune system prevents oncogenesis through 3 main mechanisms. Firstly, the immune system eliminates or suppresses viral infections to prevent virus-induced tumours, as seen in the role of Ebstein Barr virus infections in the development of early post-transplant lymphoproliferative disorder (PTLD)[6,10]. Secondly, inflammation resolution and pathogen elimination prevents the establishment of a pro-inflammatory environment conducive to tumourigenesis[6,10]. Thirdly, cells of the innate and adaptive immune system can identify and eliminate tumour cells based on the expression of tumour-specific antigens and danger signals[6,10]. Chronic immunosuppression exposure disrupts the integrity of cancer immunosurveillance. Furthermore, animal studies have suggested that tumours developing in an immunocompromised host are more immunogenic compared to an immunocompetent host, enabling tumour cells to evade immune recognition and destruction[10]. Unsurprisingly, the incidence of PTM is as high as 20% in solid organ transplant recipients after 10 years of cumulative immunosuppression exposure[6].

Potential mechanisms of carcinogenicity of individual immunosuppression drugs

Individual immunosuppression drugs may also have direct carcinogenic effects, resulting in DNA damage and gene expression changes that promote cancer progression independent of the effects of overall immunosuppression exposure.

Tacrolimus

Calcineurin inhibitors (CNI) such as tacrolimus and cyclosporine suppress T cell activation and proliferation by inhibiting interleukin-2 gene transcription (Figure 2)[11]. The reduced rate of cellular rejection and resultant improved graft and patient survival associated with tacrolimus-based immunosuppression has led to tacrolimus being the CNI of choice following solid organ transplantation[12]. However, experimental data suggest tacrolimus may promote cancer progression by creating a tumour-permissive microenvironment independent of its immunosuppressive effects.

Figure 2 Mechanism of action of commonly used immunosuppression drugs following solid organ transplantation. APC: Antigen presenting cell; IL: Interleukin; MHC: Major histocompatibility complex; mTOR: Mammalian target of rapamycin; NFAT: Nuclear factor of activated T cells.

Tacrolimus has a dose-dependent effect on the production of transforming growth factor β1 (TGF-β1), a cytokine implicated in tumour growth, metastatic spread and development of biologically aggressive cancers[13,14]. The microenvironment is further altered by TGF-β1 through inhibition of anti-tumour immune responses and promotion of extracellular matrix production and angiogenesis[15].

The direct effect of tacrolimus on tumour angiogenesis is not fully understood and may be tissue-dependent. In vivo studies have demonstrated tacrolimus enhanced lymphangiogenesis and invasion of hepatocellular carcinoma via increased vascular endothelial growth factor (VEGF)-C expression[16]. However, tacrolimus may also hinder angiogenesis through its indirect inhibition of nuclear factor of activated T cells, which has a critical role in mediating angiogenesis through its stimulation of VEGF and secreted frizzled-related protein 2[17,18]. This anti-angiogenic effect has an emerging therapeutic role in rheumatoid arthritis, breast cancer, corneal neovascularisation and hypertrophic scars[17,18].

Tacrolimus exposure may lead to alterations in gene expression that promote cancer development and progression. Tacrolimus has been found to activate the proto-oncogene, Ras, in human renal epithelial cells and renal cancer cells, contributing to renal cancer development[19]. Notably, the activation of Ras is critical for VEGF over-expression and subsequent angiogenesis[19]. Tacrolimus can also interfere with proline-oxidase and p53-mediated apoptosis, thus promoting tumour growth[20].

Experimental data on cyclosporine has similarly demonstrated its oncogenic effects through the over-expression of TGFβ and VEGF, impaired repair of radiation-induced DNA damage and promotion of apoptosis[21-25]. The shared mechanism of action between tacrolimus and cyclosporine possibly reflects a class-effect of CNIs on malignancy risk.

MPA

Mycophenolate mofetil is a key component of backbone immunosuppression following LT, allowing for CNI de-escalation or cessation and minimisation of renal and metabolic dysfunction. The active metabolite, MPA, inhibits inosine monophosphate dehydrogenase (IMPDH) which is a crucial enzyme involved in de novo guanosine nucleotide and DNA synthesis (Figure 2)[26]. This leads to the preferential depletion of lymphocytes due their dependency on de novo purine synthesis[26]. There is currently no experimental data linking MPA to increased carcinogenicity risk independent of its effects associated with overall immunosuppression. On the contrary, MPA has in vitro and in vivo anti-neoplastic properties which may confer a reduced risk of PTM.

MPA has been shown to inhibit the growth of a variety of in vivo tumour cell lines[27-31]. Upregulation of peroxisome proliferative-activated receptor gamma by MPA prevents tumour cell differentiation[32]. Reduced expression of adhesion molecules on lymphocytes and endothelial cells interferes with adhesion receptor-dependent tumour dissemination[33-36]. Furthermore, increased expression of subtypes of adhesion receptors from the β1 integrin family may induce re-differentiation of tumour cells towards a lower invasive phenotype[36]. However, some cancer types have been found to be resistant to the anti-neoplastic properties of MPA[28,37].

The anti-neoplastic properties of MPA may also have a therapeutic potential. The enzyme IMPDH, the target of MPA, is over-expressed in cancer cells[26]. Furthermore, MPA-mediated inhibition of IMPDH has been demonstrated to induce tumour cell apoptosis, however these findings are yet to be confirmed in vivo[38].

Other immunosuppression agents

Azathioprine is a purine analogue that is incorporated into cellular DNA, where it inhibits purine nucleotide synthesis and interferes with RNA synthesis and metabolism (Figure 2)[15]. It is well known that azathioprine is a risk factor for the development of PTM, in particular, non-melanoma skin cancer. Multiple studies have demonstrated the synergistic effect between ultraviolet A radiation and the azathioprine metabolite, 6-thioguanine, in the generation of mutagenic oxidative DNA damage[39,40]. The carcinogenic effects of azathioprine have limited its use in transplantation in favour of MPA.

Sirolimus and everolimus inhibit mammalian target of rapamycin (mTOR), which subsequently downregulates cyclin-dependent kinases and mRNAs required for cell cycle progression, thus preventing interleukin-2-mediated lymphocyte proliferation (Figure 2)[9]. In vivo studies have shown mTOR inhibitors precipitate tumour cell cycle progression arrest and subsequent apoptosis[41,42]. Impaired VEGF production and signalling also restricts tumour angiogenesis and metastatic spread[43-45]. Interestingly, the simultaneous administration of sirolimus in these models can reverse the pro-angiogenic effects of cyclosporine[43-45]. The potential dual immunosuppressive and anti-neoplastic properties of mTOR inhibitors has led to its increasing utilisation in the transplantation setting.

MATERIALS AND METHODS

Multiple population and cohort studies have investigated the role of tacrolimus and MPA in the development of de novo PTM, however a direct causal relationship is difficult to establish. As the two most commonly used drugs for maintenance monotherapy post-LT, the oncogenic risk profile of tacrolimus and MPA warrants further review.

Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines were utilised to shortlist relevant articles for this narrative review to minimise bias. A comprehensive literature search was conducted through MEDLINE and Embase electronic databases between 1^st January 2002 to 11th August 2022. This time period was selected to include relevant literature since the introduction and clinical use of MPA. The following terms were used, including synonyms and closely related words, as Medical Subject Headings (MeSH) and text words: “Solid Organ Transplantation”, “Tacrolimus”, “Mycophenolic Acid”, and “Carcinogenicity”. Explosion of MeSH terms, Boolean operators and truncations were also utilised throughout the search. Further articles were identified through reference lists of published systematic reviews in the area. Excluding duplicates, abstracts from 1230 records were screened by a single reviewer, whereby 31 records were deemed appropriate for full-text review. Full-text articles were assessed for eligibility based on inclusion and exclusion criteria, leaving 6 studies for inclusion in this review (Table 1, Figure 3).

Figure 3 Search strategy utilised for article selection.

Table 1 Inclusion and exclusion criteria utilised for literature search strategy.

Inclusion criteria	Exclusion criteria
Involve human solid organ transplant recipients	Presents risk data on only one of the immunosuppressant medications
Independent malignancy risk analysis related to both immunosuppressants mycophenolic acid and tacrolimus	Does not specify type of immunosuppression
Contains a group of participants exposed to tacrolimus or mycophenolic acid, exclusive of the other	Mean follow up less than one year (given the slow growing nature of malignancy)
Greater than 100 participants	Not published in English
Greater than 5 cases of malignancy	Full text not available
Randomised controlled trials and observational studies	Systematic reviews and meta-analyses

RESULTS

The 6 studies included in this review are summarised in Table 2. All studies were large population-based registries or cohort studies that analysed PTM risk in the presence or absence of tacrolimus or MPA use. No studies included data on individual drug dosages, plasma levels or duration to assess for cumulative drug exposure. There was heterogeneity amongst the studied populations in type of organ transplanted, transplantation era and immunosuppression regimens used. No studies provided a direct comparative risk of PTM with tacrolimus or MPA monotherapy.

Table 2 Summary of studies comparing tacrolimus to mycophenolic acid in solid organ transplantation.

Ref.	Transplant era	Organ	n	De novo malignancy	Time to malignancy (median)	TAC vs no TAC (OR/HR/IRR)	MPA vs no MPA (OR/HR/IRR)
Van Leeuwen et al[50], 2009	1982 to 2003	Kidney	8162	203 lip SCC; 121 lip SCC of the lower vermillion in first transplant)	6.1 yr	SCC of the lower vermillion of lip during first transplant [IRR: 2.07, 95%CI: 0.45-9.50 (P = 0.35)]	SCC of the lower vermillion of lip during first transplant [IRR: 0.85, 95%CI: 0.28-2.60 (P = 0.77)]
Caillard et al[51], 2012	1998 to 2007	Kidney ± pancreas	21351	181 PTLD; 43 graft PTLD	Not specified	Any PTLD [aHR: 0.66, 95%CI: 0.36-1.22 (P = 0.19)]; Graft PTLD [HR: 0.33, 95%CI: 0.16-0.68 (P = 0.003)a]	Any PTLD [aHR: 1.22, 95%CI: 0.74-2.02 (P = 0.44)]; Graft PTLD [HR: 0.44, 95%CI: 0.23-0.86 (P = 0.015)a]
Hsiao et al[47], 2014	2000 to 2008	Kidney	642	54 non-cutaneous malignancy	3.9 yr	HR: 1.99, 95%CI: 0.66-6.00 (P = 0.22)	HR: 1.00, 95%CI: 0.40-2.45 (P = 0.99)
Coghill et al[49], 2016	1995 to 2010	Kidney ± pancreas, and heart	2004	170 SCC	9.0 yr	Single SCC (OR: 1.11, 95%CI: 0.48-2.60)	Single SCC (OR: 0.52, 95%CI: 0.32-0.84a)
Yeh et al[46], 2020	1997 to 2011	Liver, kidney, and heart	7852	612 cutaneous and non-cutaneous malignancy	Not specified	cHR (heart): 0.6, 95%CI: 0.1-2.7; cHR (kidney): 1.5, 95%CI: 0.8-2.6; aHR (liver): 0.6, 95%CI: 0.2-1.7	cHR (heart): 1.6, 95%CI: 0.7-3.3; aHR (kidney): 1.5, 95%CI: 1.2-1.8 (P < 0.001)a; cHR (liver): 1.5, 95%CI: 0.9-2.5
Gibson et al[48], 2021	2010 to 2018	Liver, kidney ± pancreas, heart, and lung	2852	242 cutaneous malignancy	4.7 yr	IRR: 0.83, 95%CI: 0.55-1.25 (P = 0.37)	IRR: 0.78, 95%CI: 0.54-1.12 (P = 0.18)

^aP < 0.05.

aHR: Adjusted hazard ratio; cHR: Crude hazard ratio; HR: Hazard ratio; IRR: Incidence rate ratio; MPA: Mycophenolic acid; OR: Odds ratio; PTLD: Post-transplant lymphoproliferative disorder; SCC: Squamous cell carcinoma; TAC: Tacrolimus; 95%CI: 95% confidence interval.

Cutaneous and non-cutaneous malignancy

A Taiwanese population-based study evaluated risk factors for de novo cutaneous and non-cutaneous malignancy in 7852 liver, heart, and kidney transplant recipients[46]. Among 2127 liver transplant recipients, 111 (5.2%) malignancies were recorded during the mean follow-up period of 4.2 years[46]. Despite the majority of liver transplant recipients using tacrolimus (77.3%) or MPA (99.0%), neither immunosuppressant was associated with PTM risk[46].

Among 687 heart transplant patients, 31 (4.5%) de novo malignancies were reported[46]. Immunosuppression therapy was also not associated with PTM risk in this cohort[46]. However, the smaller number of malignancy outcomes may have contributed to attenuated risk estimates.

De novo malignancy was diagnosed in 470 out of 5038 (9.3%) kidney transplant recipients[46]. The use of MPA was an independent risk factor for PTM in kidney transplant recipients, compared to no MPA use [adjusted hazard ratio (HR): 1.5, 95% confidence interval (95%CI): 1.2-1.8; P < 0.001][46]. MPA exposure was also a risk factor for de novo transitional cell carcinoma (adjusted HR: 1.7, 95%CI: 1.2-2.4; P < 0.01) and renal cell carcinoma (adjusted HR: 1.7, 95%CI: 1.1-2.8; P < 0.05) in a sub-analysis of kidney transplant recipients without hypertension or diabetes as an underlying cause for renal failure[46].

On the contrary, a smaller Taiwanese population cohort study of 642 kidney transplant recipients did not demonstrate an association between MPA or tacrolimus exposure, and the development of 54 (8.4%) de novo malignancies[47]. However, the study’s primary endpoint of hospitalisation due to malignancy as the primary coded diagnosis, likely underestimated the incidence of de novo PTM from the exclusion of malignancies coded as secondary diagnoses or those diagnosed in the community.

Differences in immunosuppression regimens and cumulative exposure to individual drugs may also contribute to the conflicting findings of the aforementioned studies, however this data was not available for analysis. Additionally, lifestyle factors known to influence malignancy risk such as smoking and alcohol consumption, were not included in either study.

Cutaneous malignancy

Three studies investigated the relationship between immunosuppression and post-transplant cutaneous malignancy.

A population-based study in the United Kingdom investigated the development of post-transplant melanoma and non-melanoma skin cancers in 2852 liver, kidney, pancreas, heart, and lung transplant recipients, compared to 13527 matched controls from the general population[48]. Among 437 liver transplant recipients, 19 (4.3%) skin cancers were diagnosed during the 6.2 year median follow-up period[48]. Liver transplant recipients had the lowest incidence of skin cancer compared to other solid organ transplant recipients [Incidence rate ratio (IRR): 4.34, 95%CI: 2.48-7.58, P = 0.00], possibly reflecting lower immunosuppression requirements and relative immune privilege[48]. Neither tacrolimus nor MPA use was associated with the development of de novo cutaneous malignancy across all solid organ transplantation[48]. However, these findings are limited by small outcome numbers. Additionally, the complex interaction between immunosuppression agents and other risk factors for skin cancer including smoking status and ultraviolet light exposure was not considered.

An American study compared 170 kidney, kidney/pancreas, and heart transplant recipients with de novo cutaneous squamous cell carcinoma (SCC) to 324 matched recipient controls[49]. Risk factors such as smoking status, family history of skin cancer and personal history of pre-cancerous skin lesions were adjusted for, however the cancer group were significantly older than the non-cancer group despite matching. In azathioprine naïve patients, MPA use was associated with lower cutaneous SCC risk, independent of tacrolimus exposure (OR: 0.52, 95%CI: 0.32-0.84)[49]. Current and previous MPA use was also inversely associated with the development of multiple cutaneous SCCs (previous MPA use: OR: 0.53, 95%CI: 0.3-0.94; current MPA use: OR: 0.52, 95%CI: 0.29-0.94)[49]. Conversely, cyclosporine-naïve patients treated with tacrolimus had no significant difference in cutaneous SCC risk compared to no tacrolimus use, when adjusted for MPA exposure[49]. Although the authors considered individual immunosuppression exposure risk in the clinical context of changing multi-drug regimens, this was limited by potential recall bias associated with self-reported questionnaires used to obtain immunosuppression data.

Finally, de novo lip SCC was evaluated in a large Australian and New Zealand registry study of 8162 kidney transplant patients[50]. Mycophenolate use was associated with reduced risk of SCC of the lower vermillion of the lip in univariate (IRR: 0.28, 95%CI: 0.12-0.69, P = 0.006), but not multivariate (IRR: 0.85, 95%CI: 0.28-2.60, P = 0.774) analyses[50]. There was no difference between tacrolimus use vs no use in the risk of lip SCC of the lower vermillion (IRR: 2.07, 95%CI: 0.45-9.50, P = 0.35)[50]. Of note, the study included patients transplanted between 1982 and 2003, with less use of tacrolimus (2/121, 1.7%) and MPA (5/121, 4.1%) during this transplant era, compared to cyclosporine and azathioprine, respectively. This study was likely underpowered to draw conclusions between tacrolimus and MPA exposure and risk of SCC of the lower vermillion of the lip.

PTLD

A large population registry in France evaluated risk factors for PTLD occurrence in kidney and kidney/pancreas transplant recipients over a 10-year period[51]. Compared to 21170 control kidney transplant recipients, 327 cases of PTLD were recorded and 181 cases were included in the final analysis[51]. Tacrolimus and MPA use were not associated with overall PTLD risk, even when simultaneous kidney pancreas transplant recipients were excluded[51]. However, tacrolimus and MPA were negatively associated with graft site PTLD (tacrolimus: HR: 0.33, 95%CI: 0.16-0.68; MPA: HR: 0.44; 95%CI: 0.23-0.86), which may be attributed to fewer episodes of acute rejection and less immunosuppression exposure in this subgroup[51].

DISCUSSION

With long-term survival now commonplace following LT, there is an increasing need to improve non-hepatic health to avoid complications including metabolic derangements, renal impairment and de novo malignancy. De novo PTM accounts for approximately 16.4% of late deaths following LT[12,52]. Although immunosuppression exposure is a well-known contributor of PTM risk, there remains uncertainty regarding the carcinogenic effect of specific immunosuppression drugs, alone or in combination. This is the first narrative review that compares the relative carcinogenicity of tacrolimus and MPA in solid organ transplant recipients.

Existing in vitro and in vivo experimental data have portrayed a contrasting carcinogenic risk profile between tacrolimus and MPA. Tacrolimus promotes oncogenesis and tumour growth in its surrounding microenvironment with the activation of proto-oncogenes, production of TGF-β and inhibition of apoptosis[13,19,20]. The data on MPA is limited but suggests possible inhibition of tumour cell differentiation and prevention of vascular spread through alteration of cellular adhesion molecule expression[6]. However, there is currently no human data that directly compares the carcinogenic effects of tacrolimus and MPA in LT or other solid organ transplantation.

This review included a small number of studies that did not demonstrate a clear difference between tacrolimus and MPA in de novo PTM risk following solid organ transplantation. Our findings are in keeping with a recent systematic review and meta-analysis of kidney, liver, heart, and lung transplant recipients, whereby the risk of de novo malignancy did not differ between patients who received MPA and patients who received tacrolimus (OR: 0.88, 95%CI: 0.69-1.14, P = 0.33)[53]. However, the relationship between immunosuppression exposure and de novo PTM risk may vary based on transplant type. In liver transplant recipients, cumulative tacrolimus exposure has been associated with the development of PTM[54,55], although the high tacrolimus doses utilised in these studies are no longer aimed for in routine clinical practice. Furthermore, the conversion from CNI-based immunosuppression to MPA monotherapy post-LT results in either similar or lower rates of PTM[56,57]. Whether the reduction in PTM risk found in these studies is due to the effects of MPA or the reduction in tacrolimus exposure, is unknown. Thus, the differential carcinogenic risk profile of tacrolimus and MPA found in previous experimental studies is yet to be replicated in the clinical setting. Further clarification with large prospective studies is required.

There are inherent practical and financial difficulties in designing studies to compare the relative risk of de novo PTM between tacrolimus and MPA. Large prospective population-based studies of prolonged follow-up duration are required to ensure adequate statistical power. Variables that influence PTM risk such as age, gender, ethnicity, and smoking should be identified. However, there may be unidentifiable confounders that are difficult to capture, owing to the complex interaction between genetic, lifestyle and disease factors in oncogenesis. Population-based registries often rely on International Classification of Diseases coding for data collection, which can lead under-representation of malignancy incidence due to miscoding. Finally, longitudinal recording of drug dose, plasma levels and duration is required to capture changes in immunosuppression regimens frequently seen in routine clinical practice. The accurate calculation of cumulative immunosuppression exposure minimises drug exposure misclassification bias seen in current transplant cohort analyses that presume an unvarying drug regimen.

Immunosuppression minimisation is an important strategy to reduce PTM risk given the limited clinical data surrounding individual agents. There are currently no clear guidelines regarding immunosuppression drug choice to minimise PTM risk following LT. European LT guidelines state CNI-related de novo PTM risk may be due to dosage, and that there is no evidence to suggest MPA contributes to de novo PTM development[58]. In our centre, there is a preference for MPA, alone or in combination with everolimus, due to improved renal outcomes and experimental data suggesting higher PTM risk with tacrolimus. Overall, the choice of immunosuppression needs to be individualised based on recipient characteristics, liver disease aetiology, and alloimmune risk.

Routine cancer surveillance for all transplant recipients is recommended in addition to immunosuppression minimisation. Strict cancer surveillance strategies may lead to earlier cancer detection rates and improved non-cutaneous cancer patient survival in LT recipients[59,60]. As non-melanoma skin cancer is the leading cause of PTM in LT recipients, annual skin examinations by a dermatologist are recommended from 5 years or more after LT[5,61]. Recipients with primary sclerosing cholangitis and inflammatory bowel disease require annual colonoscopies for colorectal cancer surveillance[61]. Age and gender based cancer surveillance for all LT recipients is also recommended.

CONCLUSION

The clinical relevance of previous experimental studies on the relative carcinogenicity of tacrolimus and MPA, and its application in solid organ transplantation, is yet to be confirmed. Consequently, the choice of immunosuppressive agent to use as maintenance monotherapy in LT patients is not currently supported by a strong evidence base and remains unclear. Further studies are required to enable physicians to tailor immunosuppression regimens to minimise individual malignancy risk.

ARTICLE HIGHLIGHTS

Research background

Many liver transplant (LT) recipients are able to be maintained on long-term immunosuppressive monotherapy, most commonly with either tacrolimus or mycophenolate. In experimental studies, tacrolimus is associated with increased carcinogenicity, whereas mycophenolic acid (MPA) may have anti-neoplastic properties. However, there is minimal clinical data comparing the relative carcinogenicity of tacrolimus and MPA in LT or other solid organ transplant recipients.

Research motivation

Post-transplant malignancy (PTM) is a leading cause of late mortality in LT recipients. Thus, a clinically relevant difference in the carcinogenic risk profile between tacrolimus and MPA will affect the choice of immunosuppressive agent used as maintenance monotherapy in LT patients.

Research objectives

To determine the relative carcinogenicity of tacrolimus and MPA in solid organ transplantation.

Research methods

A systematic review was conducted using PRISMA guidelines with relevant articles published between 1^st January 2002 to 11^th August 2022 retrieved from MEDLINE and Embase databases for review.

Research results

A total of 6 studies were included in this systematic review, which did not demonstrate a clear difference between tacrolimus and MPA in the development of de novo PTM following solid organ transplantation.

Research conclusions

The relative carcinogenicity of tacrolimus and MPA, and its clinical relevance in solid organ transplantation, remains unclear.

Research perspectives

This review highlights the need for further large, population-based prospective studies to further assess the carcinogenic profiles of tacrolimus and MPA, to assist physicians in the choice of immunosuppressive agent to use as maintenance monotherapy in LT patients.