Enfortumab Vedotin-Induced Febrile Neutropenia and Hyperglycemia Successfully Treated with Multidisciplinary Treatment Including Continuous Hemodialysis Filtration and Insulin Injection in a Patient with Chemo-Resistant Metastatic Urothelial Carcinoma: A Case Report

Enfortumab vedotin (EV) is an antibody-drug conjugate adopted as a third-line agent for adult patients with metastatic urothelial carcinoma who have previously failed platinum-based therapy and a PD-1/PD-L1 inhibitor. EV showed significant overall survival benefit over conventional chemotherapy [1]. Hyperglycemia and diabetic ketoacidosis (DKA) have been reported in patients treated with EV. In a global phase III trial, 6.4% of patients experienced treatment-related hyperglycemia [1]. Hyperglycemia occurred more frequently in patients with a history of diabetes, BMI ≥30 kg/m², hemoglobin A1c ≥6.5% [1]. In the literature, several case reports suggest the association between EV and hyperglycemia, and negative outcomes such as multiple organ failure and death [2]. The Japanese subgroup analysis confirmed the consistent efficacy and safety of EV [3]; however, the present case describes the adverse effects of uncontrollable hyperglycemia and febrile neutropenia (FN) and a positive outcome achieved with multimodal therapy.

A 72-year-old Asian man with mild obesity and type 2 diabetes underwent cystectomy and ileal conduit reconstruction for muscle invasive bladder carcinoma 18 months before entering the hospital for the episode of this report. Before undergoing cystectomy, he had been first diagnosed with bladder urothelial carcinoma, high grade, pT1 by transurethral resection and followed by Bacillus Calmette-Guerin intravesical therapy. During the Bacillus Calmette-Guerin therapy, the bladder tumor had reemerged, and another transurethral resection revealed the tumor was muscle invasive urothelial carcinoma, pT2. Computed tomography showed no metastasis then and cystectomy and ileal conduit reconstruction were performed following 4 courses of gemcitabine and cisplatin chemotherapy. The pathology was urothelial carcinoma, pT3a with lymphovascular invasion. Five months after the cystectomy, metastatic urothelial carcinoma recurred in the lumbar vertebra, and pembrolizumab together with radiation therapy was started. The metastatic site remained the same size during nine courses of pembrolizumab treatment. Corporal metastasis and lumbar vertebrae metastasis had appeared after nine courses of pembrolizumab and the patient’s treatment was changed to EV 1.25 mg/kg on days 1, 8, and 15 for a 28-day cycle was started. The patient’s oncology treatment timeline is shown in Figure 1. He had received medical therapy of anagliptin 200 mg/day for diabetes mellitus when treated with EV. His fasting blood glucose was around 100–120 mg/dL during the treatment of days 1 and 8 of EV. It had risen to 226 mg/dL before day 15. The laboratory status just before day 15 of EV, which was his entrance to our hospital in this report, is presented in Table 1. The blood glucose control had been difficult regarding the hemoglobin A1c of 8.8% (Table 1). To control the high blood glucose, he was started on intensified oral treatment and insulin therapy. He then developed FN and was transferred to the intensive care unit (ICU). Despite intensified insulin therapy, his blood glucose remained high, and continuous intravenous insulin infusion (CVII) was started. The total insulin dose was stopped at a maximum of 316 units/day because of the caution for hypoglycemia which is anticipated after the reduction of the effect of EV. His blood gas showed acidosis, but anion gap (AG) did not increase, and urinary ketones were negative, which indicates that the patient did not develop DKA. The patient developed pyelonephritis and FN at the same period and he became septic 3 days after diagnosis. Under septic condition, noradrenaline was administrated for the circulatory instability due to hypovolemia and systemic inflammation. This hypovolemic and inflammatory condition also induced acute renal injury (AKI). Continuous hemodialysis filtration (CHDF) was adopted for hemodialysis with minimal circulatory fluctuation and there was also an additional intent of reducing the blood glucose levels in a sustained manner during blood purification. CHDF was performed at a blood flow rate of 100 mL/h, filtration pump flow rate of 800 mL/h, dialysate pump rate of 500 mL/h, and replacement fluid pump flow rate of 300 mL/h. Replacement fluid pump flow rates were modified accordingly. For the dialysis treatment, the membrane CH-1.8w (Toray Ind., Tokyo, Japan; polymethylmethacrylate: PMMA) was employed, which is known for polymethylmethacrylate hemofilter with an effective surface area of 1.8 m². The treatment timeline for FN is presented in Figure 2. The blood glucose level gradually fell and the total insulin dose was reduced to 30–60 units/day after CHDF treatment. The clinical course is shown in Figure 2. Nutritional management and insulin dosage adjustments were then conducted to prevent the patient from developing DKA. Eventually, the patient’s blood glucose levels stabilized with 3 units/day of insulin glargine and 5 mg/day of linagliptin, an inhibitor of dipeptidyl peptidase-4, and he was released from ICU. The patient had spent for total of 2 months in our hospital and was finally discharged. His blood glucose level stabilized after discharge from the hospital with only 5 mg/day of linagliptin without insulin and the fasting blood was around 100–150 mg/dL.

There are several reported cases of hyperglycemia, DKA, and acute kidney failure associated with EV [2, 4, 5]. The clinical outcome of the reported cases is shown in Table 2. Most of the cases exhibit hyperglycemia 9 days after the initial treatment of EV. In the phase II EV-201 study, which used EV after immune checkpoint inhibitors in chemo-ineligible patients with advanced urothelial carcinoma, grade 3–4 hyperglycemia was presented in 7% [3]. Therefore, there could be link in using both immune checkpoint inhibitors and EV in introducing hyperglycemia. Type 1 diabetes is a known side effect of immune checkpoint inhibitors, and the frequency of type 1 diabetes caused by pembrolizumab, is reported to be about 0.4% in all grades [6]. In the present case, the C-peptide immunoreactivity was 4.02 ng/mL 2 weeks after the EV treatment, and the insulin secretory capacity was maintained during the hyperglycemic period. It is not likely that the development of type 1 diabetes was caused by pembrolizumab, and it is almost clear that hyperglycemia was derived from EV treatment in the present case.

Three major adverse events of EV are hyperglycemia, peripheral neuropathy, and skin rash. These three events during clinical trials required special attention: hyperglycemia (6%), peripheral neuropathy (34%), and rash (47%) [7]. The mechanism by which EV causes hyperglycemia is unknown. A significant increase in glucose uptake in human skeletal muscle cells and a decline in insulin secretion from human islets after exposure to monomethyl auristatin E, EV’s drug component, derives hyperglycemia. This was noted in EV’s FDA application [8]. To be noted, in a case of non-diabetic patient, DKA and AKI were induced by EV and he succumbed to death even despite the treatment of continuous venovenous hemodialysis and insulin drip [2]. Obesity and preexisting diabetes are considered risk factors and patients with such factors should be monitored for hyperglycemia when EV is administered [2].

The presented patient had a BMI of 26.9 kg/m², a high HbA1c of 6.9% 2 months before EV administration, and a history of type 2 diabetes mellitus (Table 1). The blood glucose control was acceptable as it showed under 100–120 mg/dL in the beginning of EV; therefore, we decided to continue EV because of these laboratory findings. After 14 days of CHDF and respiration care, the patient blood glucose became stable and the patient finally recovered from AKI and hyperglycemia (Fig. 2). When compared to a similar case with DKA and multiple organ failure, it is suggested that the difference was the control of pneumonia which was not successful in the reported case [2, 4]. It is known that infection can cause impaired glucose tolerance and hyperglycemia; therefore, clinicians should monitor the patient with great care when an infection such as pyelonephritis or pneumonia occurs. There is also a possibility that the infection contributed to the increased insulin resistance under EV treatment.

Based on the concept of insulin resistance of EV, there is a rationale for the non-effectivity of high doses of insulin injection. We had stopped the escalated insulin dose at 316 units/day, for fear of hypoglycemia which is anticipated after the elimination of EV. The blood glucose was monitored carefully and checked every hour during CVII.

Although CHDF is not designed to treat hyperglycemia, we expect its effect to pull out glucose from the blood vessel and control the blood glucose during the hyperglycemic stage. The liberation period from CHDF of 14 days has a temporal correlation with the elimination half-life of EV of 3.4 days (Fig. 2). CVII was treated for 26 days and changed to normal diabetes therapy. Although CHDF exhibits a reduced hourly glucose extraction rate compared to intermittent hemodialysis, it is anticipated to efficaciously address elevated blood glucose levels gradually and continuously over 24 h. Upon resolution of the life-threatening hyperglycemic phase, it becomes imperative to sustain corrective measures while managing significant fluctuations in blood glucose levels. In this context, CHDF emerges as a commendable instrument for such endeavors. Since a decrease in glucose uptake into the cells leads to progression to DKA, enteral nutrition was also performed at the same time in this case. The fact that DKA did not occur in this case, compared with other cases of DKA, may have been due to appropriate nutritional management.

The present case describes a positive outcome with recovery from hyperglycemia after treatment with CHDF and CVII. After 2 months of intensive care with CHDF, CVII, and respiration care, the patient blood glucose recovered to stable condition and finally the patient also recovered from hyperglycemia and sepsis.

Hyperglycemia as an adverse event of EV, sometimes exhibits extremely high insulin resistance and causes a poor prognosis. Patients with a history of diabetes or high BMI or HbA1c are at risk for hyperglycemia and should be followed carefully. Blood sugar control and strict systemic management through CHDF and continuous insulin administration in the ICU contribute to improved prognosis.

We thank the patient and his family for supporting us by releasing his clinical data for this publication. We also thank the staff at the intensive care unit and 5th floor north ward for providing care for the patient. Last, but not least, we thank Masayuki Takeda for always sharing ideas and meaningful discussion.

Written informed consent was obtained from the patient for publication of this case report and accompanying images. No identifiable images or data are included in this report. This study was approved by the Ethics Committee of University of Yamanashi Hospital (Approval No. 2616; Yamanashi, Japan). The CARE Checklist has been completed by the authors for this case report, attached as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000540354).

All the authors declare no conflict of interest.

The authors have no funding to any research relevant to this case report.

Ayatsugu Ootsuka, Norifumi Sawada, Ryosuke Suda, Fumiakira Yano, Takuya Osada, and Yuko Otake contributed work on the paper and clinical management. Daiki Harada, Junko Goto, and Takeshi Moriguchi engaged in the intensive care and supported the patient. Tomomi Watanabe, Tadatsugu Hosokawa, Kyoichiro Tsuchiya has treated the patient for the diabetes treatment whole the time course. Hiroshi Shimura, Takanori Mochizuki, Satoru Kira, and Takahiko Mitsui participated in the preparation of the paper. All authors contributed to the work and have read and approved the final manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.