Dasatinib-induced colitis: a case report

We report a case of subclinical colitis induced by dasatinib (a BCR/ABL tyrosine kinase inhibitor) in a 61-year-old female with chronic myelogenous leukaemia (CML) in chronic phase. CML was diagnosed in late 2013. Initial breakpoint cluster region-Abelson murine leukaemia (BCR-ABL) quantitative level was 26% prior to the commencement of treatment. Dasatinib 100 mg was started in December 2013. There was an excellent response with reduction in the BCR-ABL quantitative level to 0.18% by 3 months, and major molecular response after 6 months of dasatinib therapy. After 9 months of therapy, it was noted that the patient had developed a microcytic blood film with iron deficiency; ferritin was 39 μg/L with a transferrin saturation of 10%. Physical examination was normal and she had no significant gastrointestinal symptoms or signs of gastrointestinal bleeding. A stool culture for common bacterial pathogens and parasites was negative. To investigate the patient's recent onset iron deficiency, a referral to a gastroenterologist was made and gastroscopy and colonoscopy were performed. Gastroscopy was unremarkable; however, the colonoscopy identified numerous, diminutive (<3 mm), pigmented nodules throughout the colon. Histology revealed a mild to moderate, active inflammation with cryptitis. The inflammatory infiltrate contained a mixture of lymphocytes, plasma cells, eosinophils and neutrophils. Very occasional branched crypts and borderline basal plasmacytosis were also features. Crypt abscesses were not seen. There was no ulceration or fibrosis. Pseudopyloric metaplasia was not a feature. There were no Paneth cells. Patchy surface mucin depletion was noted, probably representing regenerative changes. The differential diagnoses for the observed endoscopic features include a drug reaction, infectious colitis (though this was excluded on stool testing), nodular lymphoid hyperplasia (more common in the small bowel), primary or secondary colonic malignancy (including squamous cell carcinoma or melanoma). Inflammatory bowel disease was thought to be less likely given the unusual appearances and lack of ulceration or associated endoscopic inflammation. In addition, the atypical features and lack of risk factors meant that an ischaemic cause was excluded. These findings were thought to be related to dasatinib therapy. Despite this, as the patient had had an excellent response to dasatinib and a lack of clinically significant consequences of the colitis, the tyrosine kinase therapy was continued. She remained on dasatinib 100 mg with no clinical side effects of therapy and stable haematinics. The nodules were found at a repeat colonoscopy 3 years later (performed due to positive faecal occult blood test) and biopsied (Fig. 1). The histological findings were more prominent on this occasion, and revealed a multifocal active pan-colitis with patchy borderline basal plasmacytosis. In view of the basal plasmacytosis and previous crypt architectural distortion, inflammatory bowel disease was considered (though based on clinical history and endoscopic appearances this was thought less likely). Histologically, diverticular disease related colitis could also give this pattern of inflammation. A chronic infection was also a strong suggestion; however, stool culture was negative and selected organisms could not be found [periodic acid-Schiff (PAS) stain for amoebic organisms and CMV immunostain were negative]. Pseudomembranous colitis or antibiotic related colitis was not entertained in the histological differential diagnosis as eruptive necrosis or pseudomembranes were not seen. Ischaemic colitis was unlikely in the absence of crypt withering or capillary thrombosis. There was no vasculitis or amyloid deposition (Congo red stain did not show interstitial or vessel wall pink deposits suggestive of amyloid). There were no features of GVHD such as crypt apoptosis. Immunohistochemistry demonstrated a predominantly T-cell lymphocytic infiltrate within the lamina propria which was enriched for CD8+ compared to CD4+ (Fig. 2). There were only very occasional FOXP3+ (expressed by regulatory T-cells) lymphocytes and virtually non-existent CD56+ lymphocytes. These features are consistent with subclinical dasatinib-induced colitis. Given the lack of clinically significant symptoms and the excellent response to therapy, dasatinib therapy was continued. The Philadelphia chromosome is the result of a translocation between chromosome 9 and 22, commonly denoted as t(9; 22) (q34; q11). This results in the fusion gene, BCR-ABL1, which produces a constitutively active tyrosine kinase. The continuous, unregulated activity of this tyrosine kinase results in unchecked proliferation of haematopoietic cells and leads to the development of chronic myelogenous leukaemia (CML).1Jabbour E. Kantarjian H. Chronic myeloid leukaemia: 2018 update on diagnosis, therapy, and monitoring.Am J Hematol. 2018; 93: 442-459Crossref PubMed Scopus (220) Google Scholar Dasatinib is a second-generation tyrosine kinase inhibitor (TKI) which is approved as a first line treatment for CML.2Hochhaus A. Saussele S. Rosti G. et al.Chronic myeloid leukaemia: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up.Ann Oncol. 2017; 28: iv41-iv51Abstract Full Text Full Text PDF PubMed Scopus (212) Google Scholar Dasatinib is active against the ABL1 domain, but also has efficacy against the Src family kinases (Lck, Yes, Lyn and Fyn), c-KIT and platelet derived growth factor receptor-β (PDGFR-β). It has been shown to be effective in cases of resistance or intolerance to imatinib, due to more potent BCR-ABL antagonism. In first line therapy, dasatinib use is associated with faster and deeper levels of response compared with imatinib, with high levels of progression free, and overall, survival.3Cortes J.E. Saglio G. Kantarjian H.M. et al.Final final 5-year study results of DASISION: the dasatinib versus imatinib study in treatment-naive chronic myeloid leukemia patients trial.J Clin Oncol. 2016; 34: 2333-2340Crossref PubMed Scopus (556) Google Scholar However, the activity of dasatinib against a broader range of targets compared with imatinib may contribute to its side effect profile. Five year follow up of the phase III DASISION trial confirmed the safety and efficacy of dasatinib as first line therapy in CML.3Cortes J.E. Saglio G. Kantarjian H.M. et al.Final final 5-year study results of DASISION: the dasatinib versus imatinib study in treatment-naive chronic myeloid leukemia patients trial.J Clin Oncol. 2016; 34: 2333-2340Crossref PubMed Scopus (556) Google Scholar However, gastrointestinal tract side effects have been noted with dasatinib therapy and it has been shown to cause occult gastrointestinal bleeding in 2–12% of cases.4Quintas-Cardama A. Kantarjian H. Ravandi F. et al.Bleeding diathesis in patients with chronic myelogenous leukemia receiving dasatinib therapy.Cancer. 2009; 115: 2482-2490Crossref PubMed Scopus (105) Google Scholar,5Perdigoto D. Lopes S. Portela F. et al.Dasatinib-induced colitis in a patient with chronic myelogenous leukaemia.GE Port J Gastroenterol. 2018; 25: 198-200Crossref PubMed Scopus (4) Google Scholar Dasatinib affects platelet activation and aggregation (by inhibition of Src family kinases), as well as impairing the ability of megakaryocytes to form pro-platelets, exacerbating the patient's inability to achieve haemostasis.6Yim E. Choi Y.-G. Nam Y.-J. et al.Dasatinib induces severe haemorrhagic colitis in a patient with accelerated phase of chronic myelogenous leukaemia.Korean J Intern Med. 2018; 33: 446-448Crossref PubMed Scopus (3) Google Scholar A recent prospective study investigated the frequency of dasatinib-induced colitis and the screening efficacy of faecal occult blood tests (FOBT).7Nishiwaki S. Maeda M. Yamada M. et al.Clinical efficacy of fecal occult blood test and colonoscopy for dasatinib-induced hemorrhagic colitis in CML patients.Blood. 2017; 129: 126-128Crossref PubMed Scopus (11) Google Scholar Ten of 18 patients treated with dasatinib had a positive FOBT and went on to have colonoscopy and mucosal biopsy. Six of these patients (33%) had dasatinib-induced colitis and only one patient still had a positive FOBT when dasatinib was withheld. Another study reviewing the literature identified only 13 articles (10 case reports, 2 single centre case series and 1 multicentre series) reporting colitis directly associated with dasatinib therapy.8Shanshal M. Shakespeare A. Thirumala S. et al.Dasatinib-induced T-cell mediated colitis: a case report and review of the literature.Acta Haematol. 2016; 136: 219-228Crossref PubMed Scopus (10) Google Scholar However, this may be a result of under-reporting of this condition rather than a true reflection of its prevalence. There have been various theories postulated as to the mechanism of dasatinib-induced colitis. It has been hypothesised that dasatinib reduces immune tolerance.6Yim E. Choi Y.-G. Nam Y.-J. et al.Dasatinib induces severe haemorrhagic colitis in a patient with accelerated phase of chronic myelogenous leukaemia.Korean J Intern Med. 2018; 33: 446-448Crossref PubMed Scopus (3) Google Scholar Dasatinib interrupts the Src family kinase signalling pathway and thereby reduces the number of regulatory T cells in the bowel. In this way, dasatinib limits host natural immune tolerance to commensal intestinal microflora.6Yim E. Choi Y.-G. Nam Y.-J. et al.Dasatinib induces severe haemorrhagic colitis in a patient with accelerated phase of chronic myelogenous leukaemia.Korean J Intern Med. 2018; 33: 446-448Crossref PubMed Scopus (3) Google Scholar,9Aldoss I. Gaal K. Al Malki M.M. et al.Dasatinib-induced colitis after allogeneic stem cell transplantation for Philadelphia chromosome-positive acute lymphoblastic leukaemia.Biol Blood Marrow Transplant. 2016; 22: 1900-1906Abstract Full Text Full Text PDF PubMed Scopus (11) Google Scholar This allows commensal microflora to become pathogenic and cause the colitis that we associate with dasatinib. In our case the targeted colonic biopsies showed a primarily T cell infiltrate, with an increased CD8+:CD4+ ratio. In addition, there was a relative lack of regulatory T cells and NK cells, as evidenced by the sparsity of FOXP3 staining cells and absence of CD56 staining cells. The lack of regulatory T cells and reduced numbers of NK cells are in keeping with the presumed mechanism of dasatinib-induced colitis. Dasatinib may cause a drug-induced, immune-mediated colitis in susceptible patients. This seems to be a result of the drug's broad inhibition of off-target kinases involved in regulatory T cell and NK cell proliferation, causing impaired immune tolerance. The mechanism of immune dysregulation appears to be different from the inflammatory bowel disease model, where FOXP3+CD4+ T cells predominate. Dasatinib-induced colitis is a relatively uncommon phenomenon and reverses with drug cessation. However, in cases of milder colitis or those with limited alternative therapeutic options, continuation of dasatinib is acceptable. We report a case of subclinical dasatinib-induced colitis, to add to the developing body of literature relating to this pathological process. The authors state that there are no conflicts of interest to disclose.