Incidence and outcome of pancreatitis in children and young adults with acute lymphoblastic leukaemia treated on a contemporary protocol, UKALL 2003

Acute Pancreatitis is a well-known complication of the treatment of childhood acute lymphoblastic leukaemia (ALL). Pancreatitis is associated with significant complications, such as multiorgan failure and pseudocysts. The incidence of pancreatitis following ALL treatment is 7–18% (Raja et al, 2012), and this variation is due to differences in the definition of pancreatitis and variations in the doses of anti-leukaemic medications that cause pancreatitis. Whilst numerous anti-leukaemic medications have been reported to cause pancreatitis, the most important is Asparaginase. The development of pancreatitis is one of the commonest causes for stopping Asparaginase, because re-exposure is associated with recurrence. In contrast, stopping scheduled Asparaginase treatment because of previous pancreatitis has been linked to an increased relapse rate (Silverman et al, 2001). Recently, recommendations for the reintroduction of Asparaginase based on clinical criteria were suggested in attempt to minimize the risk of relapse without compromising safety (Raja et al, 2012). We therefore conducted a retrospective review of the clinical risk factors and outcome of pancreatitis on the recently closed UKALL 2003 trial. UKALL 2003 opened to patients aged 1–18 years with previously untreated ALL in October 2003. The upper age was increased to 20 years in September 2006 and 25 years in June 2008. This analysis includes the 3101 eligible patients who had entered the trial by 31 May 2011. Follow up was to 31 October 2011 with median follow up of 3 years, 10 months (range 1·5 months to 8 years, 1 month). Patients with pancreatitis were identified by the investigator-generated serious adverse report from the participating centres. Further data was obtained from directed questionnaires relating to the event. All of the questionnaires were returned. Pancreatitis was diagnosed according to the Atlanta criteria (Bradley, 1993), which include at least two of the following three features – clinical presentation resembling pancreatitis, serum amylase and/or lipase >3 times the upper limit of normal and characteristic imaging findings of acute pancreatitis. The severity of the episode was graded according to the Common Terminology Criteria for Adverse Events, version 4.03 (http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_8.5x11.pdf). Grade 3 was defined as severe pain, vomiting, medical intervention required (e.g. analgesia) and grade 4 was life threatening or urgent intervention required. Evaluated outcomes measures included event-free survival (EFS), overall survival (OS) and complications following pancreatitis. In view of the potential for recurrence, the trial recommendation was to avoid re-exposure to further Asparaginase, following pancreatitis. Details of the trial protocol and methods are shown in Fig 1A and in the supplementary information. The incidence of acute pancreatitis was 1·5%, and was 0·6%, 2·0% and 3·0% for treatment regimens A, B and C respectively. Whilst several factors were significantly associated with the development of pancreatitis on univariate analysis (Table 1A), only treatment regimen given, (with increasing intensity of treatment and asparaginase dose) was associated with higher odds of developing pancreatitis and remained significant on multivariate analysis (Table 1B). There was a suggestion that increasing age may also be important but models including this variable did not reach statistical significance. The CTCAE grade of pancreatitis was grade 3 in 73% and grade 4 in 27% of cases. Pancreatitis was more common in Asparaginase-containing blocks versus nonAsparaginase containing blocks (83% vs. 17%; P < 0·0001; Fig 1B). A median of 3 doses of Asparaginase was received (cumulative dose −3000 iu/m2) before the onset of pancreatitis (range 1–10). The median interval between receiving the last Asparaginase dose and developing pancreatitis was 10 d (range 3–34). Six cases had further episodes of pancreatitis (recurrence rate of 13%). Four patients with Asparaginase-associated pancreatitis – were re-exposed to Asparaginase. Of these, one case had recurrent pancreatitis with severe pancreatitis on both occasions (grade 4). There was significant morbidity (29% required admission to the intensive therapy unit, 25% developed pseudocysts, 13% developed diabetes mellitus, 8% pleural effusions and 2% renal failure). As the majority of patients were not re-exposed again to Asparaginase, most received half their scheduled Asparaginase [median scheduled Asparaginase dose received in the pancreatitis cohort was 50% (5000 iu/m2); range 17–100% (2–12 000 iu/m2)]. Despite these omissions, there was no significant difference in EFS and OS between those who did and did not develop pancreatitis (5-year EFS in the pancreatitis cohort 84%; 95% confidence interval (CI): 69–99% vs. 87%; 95%CI: 86–89% in the non-pancreatitis group; Odds ratio for EFS in the pancreatitis cohort compared to no pancreatitis = 1·36 95% CI: 0·50–3·74 P = 0·5; 5 year OS in pancreatitis cohort 92%; 95% CI: 82–100% vs. 92%; 95% CI:90–93% in non pancreatitis group; Odds ratio for OS in the pancreatitis cohort compared to no pancreatitis was 1·26 95% CI:0·36–4·45 P = 0·6). There were no deaths as a result of pancreatitis. The incidence of pancreatitis in our analysis was lower than previously reported in other studies. Potentially, differences in the incidence may reflect differences in the criteria used to diagnose pancreatitis. However in order to make definitive comparisons on the incidence of pancreatitis between different leukaemia protocols, a consensus classification is required on the definition on pancreatitis. As this does not currently exist, we used the Atlanta criteria to diagnose pancreatitis, as recently recommended. (Raja et al, 2012). An additional reason for the difference in pancreatitis seen on UKALL 2003 and other series may reflect differences in the cumulative dose of Asparaginase used. Thus Regimen A/B patients were exposed to a relatively small amount of amount of Peg-Asparaginase (3000 - 4000 iu/m2) and Regimen C received 12 000 iu/m2. In contrast, the Dana Farber Cancer Institute ALL 91-01 Consortium used a higher cumulative dose of Peg- Asparaginase (30 000 iu/m2) (Kearney et al, 2009). Asparaginase dose intensity is however, not the only factor likely to be important in the pathogenesis of pancreatitis. Regimen B patients had a three-fold higher incidence of pancreatitis compared to Regimen A despite having a similar Asparaginase exposure. This suggests that anthracycline induction and the additional cylophosphamide/cytarabine exposure in the Berlin-Frankfürt-Münster consolidation in Regimen B may be risk factors for the development of pancreatitis. In our series, Asparaginase withdrawal due to pancreatitis did not affect EFS and OS, though the overall numbers of cases with pancreatitis was small. However, in the current UKALL 2011 trial, we continue to recommend that Asparaginase be withheld following pancreatitis. Our analysis demonstrates that increasing treatment intensity and Asparaginase exposure are the biggest risk factors for developing pancreatitis. The authors are grateful to Professor Josef Vormoor for his comments. RW, SD, PI and JS collected data and helped write the manuscript. RW performed the statistical analyses. AJ, NG and SS supervised the writing of the manuscript. The authors declare no conflict of interest. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.