Search in the Abstract Database

Search Abstracts 2011

* = Presenting author

P205. The relationship among 6-TGN level, therapeutic effect and myelotoxicity of azathioprine: A study in a cohort of Chinese IBD patients

X. Gao

The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China

Aim: Evaluate the relationship among 6-TGN level, therapeutic effect and myelotoxicity of AZA in Chinese IBD.

Methods: AZA was administrated on fixed dose 2 mg·kg−1·day−1 throughout the study. Therapeutic effect and myelotoxicity were recorded. 6-TGN was detected at week 2, 4, 8, 12, 24, 48. The end-points were: fulfilling AZA treatment for 48 weeks, discontinuation of AZA because of myelotoxicity or treatment failure.

Results: 86 patients completed the study, 70 CD and 16 UC. 14 patients developed myelotoxicity. Steady state of 6-TGN was usually reached at the 4th week. The immediate 6-TGN was higher in patients with myelotoxicity than the maximum level in patients without myelotoxicity (P = 0.000). 6-TGN was the only risk factor significantly associated with myelotoxicity (OR = 54.4, 95% CI 2.5–1183, P = 0.01). The AUC for 6-TGN at points of myelotoxicity was 0.78 (95%CI 0.66–0.89, P = 0.00). A sensitivity of 50.0%, specificity of 86.6%, positive predictive value (PPV) of 17.5% and negative predictive value (NPV) of 96.8% were observed at the cut off level 420 pmol/8×108 RBC. Because 6-TGN <420 pmol/8×108 RBC at points of myelotoxicity were observed in all five patients treated over 12 wks, we deduced that the safety threshold was just suit to judge myelotoxicity happening during 12 wks (sensitivity 77.8%, specificity 84.7%, PPV 24.1%, NPV 98.4%). Of 11 UC patients, 5 were classified as treatment success and 6 as failure. 6-TGN was not significant different between two groups. Of 50 CD patients, 34 were classified as treatment success and 16 as failure. 6-TGN was significantly higher in success group (P < 0.05) and was the only factor significantly associated with therapeutic effect in CD patients. The AUC for mean 6-TGN in CD with treatment success was 0.71 (P = 0.02). The cut off 6-TGN level was 230 pmol/8×108 RBC (sensitivity 82.4%, specificity 62.5%, PPV 82.4% and NPV 62.5%). The increasing chance to get treatment success was observed in CD with 6-TGN ≥230 pmol/8×108 RBC (OR = 7.78, 95%CI 2.03–29.77). Of the 261 samples, 26 co-administrated with 5-ASA (3 g mesalazine daily) had higher 6-TGN than 235 samples on AZA alone (P = 0.000). 6-TGN of 51.9% samples on AZA mono-therapy fell in the range between effective threshold and security threshold and 12.8% above security threshold. 6-TGN of 53.8% samples co-administrated with 5-ASA was above security threshold. The frequency of myelotoxicity was higher in 12 patients co-administrated with 5-ASA than in 74 patients with AZA alone (P = 0.032).

Conclusions: 6-TGN was the only factor associated with myelotoxicity. The security threshold of 6-TGN was 420 pmol/8×108 RBC. The sensitivity, specificity, PPV and NPV for predicting myelotoxicity were 77.8%, 84.7%, 24.1% and 98.4% during during 12wks, but myelotoxicity was unpredictable after the 12th week. 6-TGN was the only factor associated with therapeutic effect of AZA in CD. The effective threshold of was 6-TGN ≥230 pmol/8×108 RBC. The frequency of myelotoxicity was higher in patients co-administrated with 5-ASA than patients on AZA alone.