Safety and adverse effect profile of colchicine in renal impairment: a systematic review of randomised trials

Implication for health policy/practice/research/medical education: There is no robust clinical research substantiating the association between colchicine toxicity and renal impairment. While this may encourage practices to be relaxed, it should continue to be prescribed cautiously in patients with poor renal function until further data becomes available. Please cite this paper as: Yaxley J, Yaxley W, Scott T. Safety and adverse effect profile of colchicine in renal impairment: a systematic review of randomised trials. J Renal Inj Prev. 2020; 9(4): e28. doi: 10.34172/jrip.2020.28. R ev ie w


Introduction
Colchicine is an anti-inflammatory drug widely used for a number of medical conditions. Side effects are not uncommon, most frequently including diarrhoea, nausea and abdominal pain. Caution and dose reduction or avoidance are advised in patients with renal impairment because of a presumed increased risk of side effects. While the recommended colchicine dose varies depending on the clinical indication and reference text used, product packaging uniformly advises reduced dosing in patients with a creatinine clearance less than 30 mL/min (1). However, the adverse effect profile of colchicine when prescribed in patients with acute renal impairment or chronic kidney disease (CKD) has never been systematically evaluated in the clinical setting. These practices may be unwarranted and result in reduced colchicine treatment efficacy. The aim of this report was to examine the adverse events of colchicine reported in randomised clinical trials when used in patients with renal impairment.

Study design
We conducted a systematic literature search of the MEDLINE and Griffith University Library electronic databases. All papers dated from inception up to and including March 3 rd 2020 were considered for review. We limited our search to randomised clinical trials. We included all randomised studies investigating oral colchicine use for any indication in any population, provided a non-colchicine comparator arm was included. Details regarding colchicine dose and duration were not examined. Only papers written in English were considered. Other inclusion criteria we required were at least some disclosure of adverse side effects, and a subgroup analysis of patients with renal impairment, particularly those with CKD. Renal impairment was defined in our review as any study groups with a documented glomerular filtration rate (GFR) less than 60 mL/min/1.73 m 2 , regardless of aetiology or acuity, or those classified by the respective article authors as having renal disease of any type. Any side effects were evaluated but we were particularly interested in gastrointestinal or renal toxicity.
The following entries were utilised for our search strategy; colchicine AND (randomised OR controlled OR placebo). The titles and abstracts of search hits were screened for relevance, with potentially suitable trials selected and duplicates removed independently by the authors. The bibliographies of included studies were also examined for any relevant titles. Discrepancies in the list of articles retrieved for inclusion were discussed among the authors and consensus reached. Data was extracted and tabulated and a meta-analysis attempted, but this was not completed owing to an inadequate study population.
Finally, a quality assessment of the evidence was undertaken using the Oxford quality scoring system and the GRADE Working Group guidelines (Table 1) (2,3). High quality evidence was considered data in which further evidence would be unlikely to change our confidence in the effect. Moderate quality data was evidence for which our confidence could be potentially modified by future research, and poor quality data was evidence for which our confidence in the effect is low and easily modifiable. This method of assessing the literature allowed objective interpretation of the strength of evidence.

Results
Our literature search identified a total of 7707 records.
The titles and abstracts of search hits were screened for relevance and duplicates removed, after which 68 articles were judged potentially suitable for review ( Figure 1) . Full texts were retrieved and cross-checked against our inclusion criteria. After our systematic review only a single randomised trial was ultimately deemed to meet inclusion criteria and address our research question (71). After some deliberation all authors were in agreement regarding the search results and included articles. Of the remaining 67 studies, all were primarily excluded because of absent subgroup data for a renal impairment cohort . Additionally, a large number of papers did not describe drug adverse effects.

Study characteristics
Our single included study was a randomised trial of 3 induction treatment regimens for biopsy-confirmed amyloidosis (71). There were 220 patients randomised into 3 arms of treatment; colchicine monotherapy, colchicine in combination with melphalan and prednisone, or melphalan and prednisone. The dosing regimens were complex and in the colchicine arms included colchicine at doses of 600 mcg twice daily, which was gradually uptitrated to a maximum tolerated dose ( Table 2). There were 72 patients in the colchicine group, 71 patients in the triple therapy group, and 77 patients in the prednisone and melphalan combination group.
The included trial was designed to investigate survival outcomes and the subjects were stratified according to principle organ involvement, one of which was renal disease. Other subgroups not germane to our current review were cardiac involvement, neurological disease, hepatopathy, and a miscellaneous group. Patients were assessed every three weeks to assess for adverse drug reactions out to a follow-up time of 1 year.
The renal disease subgroup referred to any patients with amyloidosis on renal biopsy with either normal or slightly abnormal kidney function and no other significant organ involvement, nephrotic syndrome, or advanced renal impairment defined as a serum creatinine greater than 177 μmol/L.

Study quality
The only eligible study was a single-centre trial that randomized patients into 3 study arms. There was no outline on the randomisation or blinding process and there was no placebo cohort. An outline of the patient recruitment procedure was absent except to say that all subjects had amyloidosis confirmed on target organ biopsy. Patient demographics were not explained except for a median age of 64 years. It was stated that some patients were not naïve to treatment, although the therapies and clinical background were not elucidated. It is unknown what portion of subjects were taking colchicine prior to enrolment. It was stated that prescribed drug doses were titrated to maximum tolerated levels however the precise dosing information was not disclosed and therefore could have been very inconsistent. There is no disclosure of the rates of deviation from or crossover between study arms, though there were no dropouts.
Because of a paucity of data, methodological flaws and the possibility of bias, there was unanimous agreement by the authors that the level of evidence obtained in our literature search was low or very low. The Jadad score using the Oxford tool was 1/5, indicating low data quality (2).   this paper, of whom 68 received colchicine therapy. Few adverse effect data was outlined for this subgroup. There was no difference in the incidence of dialysis-dependent renal failure, the time until dialysis, or the incidence of liver function test derangement across all groups. The incidence of dialysis was significant at 20.5% (41 patients) across all groups, which was presumably due to the natural history of renal amyloidosis rather than an adverse outcome of colchicine although this was not clarified. The actual frequency of abnormal liver tests was not quantified. We found no exploration of the typically expected adverse events such as leukopenia, thrombocytopenia or diarrhoea in the renal disease subgroup.

Discussion
Colchicine is an anti-inflammatory medication, believed to act through inhibition of neutrophil migration, chemotaxis, adhesion and phagocytosis in inflamed tissues (1). It is used frequently in immune diseases such as gout, pericarditis, familial Mediterranean fever and amyloidosis. Colchicine and its metabolites are predominantly excreted in faeces, while undergoing approximately 15% renal clearance (72). In experimental studies the colchicine half-life is reported to increase from approximately 4 hours in healthy subjects to 20 hours in patients with endstage renal failure (72). Consequently, the manufacturer recommends a dose reduction in patients with renal impairment due to the risk of drug accumulation and side effects (1,72). The commonest reported adverse effects are gastrointestinal symptoms, peripheral neuropathy and myopathy, and cytopaenia (1,71). It is the authors' experience that in clinical practice colchicine is generally prescribed cautiously in patients with renal impairment. This may reduce colchicine's efficacy in that context, and although this is consistent with general guidelines the safety and tolerability of colchicine when used in this patient population has not been systematically evaluated in the literature. After a systematic review of the literature, we were unable to find data in randomised studies demonstrating an increased incidence of adverse effects with kidney disease. Only a single randomised clinical trial by Kyle and colleagues was identified which in some way addressed this research question (71). In their cohort of patients with renal disease secondary to amyloidosis, defined as nephrotic syndrome or raised creatinine without extra-renal manifestations, colchicine was not associated with an elevated risk of dialysis, reduced dialysis-free survival, or deranged liver tests. This finding should be interpreted with caution given lack of a placebo group and the paucity of information provided about the medical context of the patients' renal background. No data were provided for adverse effects such as diarrhoea, nausea, or cytopaenia. It is important to recognise that an absence of clinical data demonstrating increased adverse events when colchicine is used in acute kidney injury or CKD does not equate to evidence of safety or tolerability. There are sound pharmacokinetic and physiological arguments for prescribing colchicine carefully when creatinine clearance is reduced. As mentioned, pharmacologic research confirms that the half-life of colchicine increases with falling creatinine. There are many observational reports of colchicine toxicity attributed to renal insufficiency (73,74). Conversely, there are also large cohort studies which have not identified an association between kidney disease and colchicine toxicity (75).
There is no specific guidance for the reporting of systematic literature reviews in which no studies are retrieved meeting the inclusion criteria. These so-called "empty reviews" may still be a source of useful information. We performed rigorous analysis of all high-level evidence available and have ensured diligent documentation of all excluded trials in our reference list. Our systematic search discovered only a single eligible article, rendering meaningful conclusions elusive. Furthermore, that paper is itself not without methodologic flaws. Although follow up was excellent, it was conducted at a single hospital with unknown patient demographics, and previous treatments and the patients' medical comorbidities were not propensity-matched. The predominant adverse event mentioned, being the need for dialysis, was arguably a measure of treatment outcome rather than a true side effect.
Based on two separate assessment tools, the quality and certainty of evidence gathered in our review must be graded as low or very low. It is not known whether loosening our inclusion criteria would have been helpful, or would have confused the data further and lead to greater dilution in the strength of evidence.

Conclusion
This is the first systematic review examining the adverse effect profile of colchicine in renal insufficiency in randomised trials. Although our literature review was comprehensive, evidence was scant and data quality was poor. It is widely recommended that colchicine dose be reduced in patients with acute or chronic renal impairment due to an increased risk of drug accumulation and side effects. In our study, we failed to identify any robust clinical research demonstrating this association. Although an absence of evidence does not definitively confirm drug safety, based on our study clinicians can probably prescribe colchicine with some reassurance in patients with kidney disease. Further research is warranted before strong recommendations to change practice can be supported.

Authors' contribution
JY was the principle investigator of this study. JY and TS were included in preparing the concept and design. JY and WY revised the manuscript and critically evaluated the intellectual contents. All authors participated in preparing the final draft of the manuscript, revised the manuscript and critically evaluated the intellectual contents. All authors have read and approved the content of the manuscript and confirmed the accuracy or integrity of any part of the work.

Conflicts of interest
There are no conflicts of interest to declare.

Ethical considerations
Ethical issues (including plagiarism, data fabrication, double publication) have been completely observed by the authors.

Funding/Support
None.