程序性死亡受体1抑制剂相关1型糖尿病的临床特点分析

Clinical characteristics of programmed death 1 inhibitors associated with type 1 diabetes mellitus

目的分析程序性死亡受体1(PD-1)抑制剂相关1型糖尿病(T1DM)患者的临床特点。方法对2020年5月至2022年12月在北京市顺义区医院内分泌科住院的5例PD-1抑制剂引起T1DM患者的临床资料进行分析。收集患者的一般临床资料、实验室检查结果,追踪4例新发糖尿病患者接受PD-1抑制剂治疗期间、T1DM发生前的血糖监测结果。记录患者住院期间的诊治方案、观察转归及出院后进行随访的血糖及C肽等结果。结果 5例患者中男4例、女1例,中位年龄63(53~71)岁,其中暴发性T1DM患者4例。4例以糖尿病酮症酸中毒(DKA)症状首次就诊。接受PD-1抑制剂第1次治疗到发生T1DM的中位时间为232 d(82~288 d),中位周期为10(2~14)个周期,其中3例分别在PD-1抑制剂治疗中断后44、40、65 d发生了DKA,期间未监测血糖。5例患者在诊断T1DM时的中位血糖为28.83(25.16~43.24)mmol/L,糖化血红蛋白为8.5%(6.8%~9.7%),空腹C肽为<0.02(<0.02~0.37)ng/ml,2 h C肽为0.06(<0.02~0.68)ng/ml,尿酮体阳性++~+++。胰岛自身抗体和甲状腺相关抗体阳性各2例。4例新发糖尿病患者,共完成了26个治疗周期,其中21个治疗周期前检测了空腹血糖,中位血糖为5.72(5.09~6.97)mmol/L。9个治疗周期后7 d内复查了空腹血糖,中位血糖为7.22(5.88~8.95)mmol/L。5例T1DM患者均救治成功,住院10~19 d。出院后1例采用胰岛素泵、4例采用基础+餐时胰岛素治疗,胰岛素用量为每日0.53~0.77 U/kg。4例患者出院后36~49 d复查C肽均低至检测水平下限。5例患者均遵循出院降糖方案控制血糖,血糖波动在6~13 mmol/L。结论 PD-1抑制剂可导致T1DM,多表现为暴发性T1DM,常以DKA起病,即使停止PD-1抑制剂治疗后仍可发生,胰岛β细胞破坏不可逆转,需要采用胰岛素进行治疗。为了早期识别,及时救治,需要在PD-1抑制剂治疗过程中以及停止治疗后持续血糖监测。

Objective To analyze the clinical characteristics of patients with programmed death receptor 1(PD-1)inhibitor associated with type 1 diabetes mellitus(T1DM).Methods Clinical data from 5 patients with T1DM induced by PD-1 inhibitor who were hospitalized at the Department of Endocrinology of Beijing Shunyi Hospital from May 2020 to December 2022 were analyzed.General clinical data and laboratory examination results were collected.The blood glucose monitoring results of 4 patients with new-onset diabetes were followed during PD-1 inhibitor treatment and before the onset of T1DM.The diagnosis and treatment plan during hospitalization,monitoring of the prognosis,and follow-up blood glucose and C-peptide results after discharge were recorded.Results There were 5 patients,4 men and 1 woman,with a median age of 63(53 to 71)years,including 4 patients with fulminant T1DM.Four patients presented with diabetic ketoacidosis(DKA)for the first time.The median time from first PD-1 inhibitor treatment to the onset of T1DM was 232(82 to 288)days,and the median number of cycles was 10(2 to 14).Three out of five patients developed DKA without blood glucose monitoring at 44,40,and 65 days after discontinuing PD-1 inhibitor treatment,respectively.At diagnosis of T1DM,the median blood glucose was 28.83(25.16 to 43.24)mmol/L,glycosylated hemoglobin was 8.5%(6.8%to 9.7%),fasting C-peptide was<0.02(<0.02 to 0.37)ng/ml,2 h C-peptide was 0.06(<0.02 to 0.68)ng/ml,and urine ketone body showed positive(++to+++).Islet-autoantibodies were positive in two patients,and thyroid-related antibodies were positive in two patients.Four patients with new-onset diabetes completed a total of 26 treatment cycles,including 21 cycles in which pre-treatment fasting blood glucose was measured,and the median blood glucose was 5.72(5.09 to 6.97)mmol/L.Fasting blood glucose was measured within 7 days after 9 treatment cycles,with a median blood glucose of 7.22(5.88 to 8.95)mmol/L.All patients with T1DM were successfully treated with a hospital stay of 10 to 19 days.After discharge,1 patient received insulin pump therapy,while the other 4 patients received basal plus meal insulin therapy,with the insulin doses ranging from 0.53 to 0.77 U/kg per day for all patients.Four patients reexamined C-peptide which decreased to the lower limit of detection between 36 and 49 days after discharge.All 5 patients adhered to the discharge hypoglycemic regimen to control blood glucose,which fluctuated between 6 and 13 mmol/L.Conclusions PD-1 inhibitors can cause T1DM,predominantly presenting as fulminant T1DM.DKA is often the first manifestation and can occur even after discontinuation of PD-1 treatment,as the destruction of pancreaticβ-cells appears to be irreversible and requires insulin therapy.Therefore,continuous blood glucose monitoring is required during PD-1 inhibitor treatment and after discontinuation of treatment for early identification and timely intervention.