Igor Alexander Harsch1, Peter Christopher Konturek2




The therapy with immune checkpoint inhibitors (Programmed cell death-1 and programmed cell death-ligand-1 inhibitors) is a novel and promising approach in cancer treatment. The mode of action can cause serious adverse advents, mainly immune-related ones. The case of a 65 year old caucasian female is reported. She suffers from hepatocellular carcinoma since 2012. She underwent several surgeries, was treated with sorafenib and had transcatheter arterial chemoembolizations. As part of an individual healing trial, the patient was treated with nivolumab every 2 or 3 weeks since September 2017 and has no recurrence of the tumor ever since. Eight months after onset of the therapy she rapidly developed an insulin-dependent diabetes with ketoacidosis. She had no previous history of diabetes. She had no type 1 diabetes-related autoantibodies or a common HLA-genotype associated with a higher risk to develop type 1 diabetes mellitus. This adverse event during therapy with an immune checkpoint inhibitor is yet reported to be rare. Cases with a rapid-onset diabetes with ketoacidosis have also been reported under treatment with other immune checkpoint inhibitors. The pathomechanism is not clear yet and which patients are at an elevated risk. An increasing use of immune checkpoint inhibitor therapy can be expected. This may cause a concurrent increase of the side effect diabetes. Glucose monitoring and sensitizing the patients for this adverse event seems reasonable.



Terapia inhibitorami immunologicznego punktu kontrolnego (inhibitory receptora programowanej śmierci komórki 1 i jego ligandów, ang. programmed cell death-1 and programmed cell death-ligand-1 inhibitors) stanowi nowatorskie i obiecujące podejście w leczeniu chorób nowotworowych. Jednakże taki mechanizm działania może być przyczyną poważnych zdarzeń niepożądanych, głównie na podłożu immunologicznym. W artykule przedstawiono opis przypadku 65-letniej kobiety rasy kaukaskiej, u której w 2012 roku rozpoznano raka wątrobowokomórkowego. Pacjentkę dotychczas poddano kilku operacjom, leczeniu sorafenibem, a także przezcewnikowej dotętniczej chemoembolizacji. Jako część indywidualizacji terapii, u pacjentki zastosowano niwolumabem, który jest podawany co 2-3 tygodnie od września 2017 roku. Od tamtej pory nie stwierdzono wznowy nowotworu. Jednakże osiem miesięcy po rozpoczęciu terapii u pacjentki szybko rozwinęła się insulinozależna postać cukrzycy przebiegająca z kwasicą ketonową. Wcześniej pacjentka nie chorowała na cukrzycę. Nie stwierdzono także u niej obecności autoprzeciwciał związanych z cukrzycą typu 1 ani genotypu powszechnych alleli HLA związanych z wyższym ryzykiem rozwoju cukrzycy typu 1. To działanie niepożądane związane z terapią inhibitorem immunologicznego punktu kontrolnego jest rzadko opisywane. Jednakże przypadki nagłego rozwoju cukrzycy z kwasicą ketonową zaobserwowano także w przypadku leczenia innymi inhibitorami immunologicznego punktu kontrolnego. Patomechanizm obserwowanego zjawiska nie jest jasny, jak też nie wiadomo, którzy pacjenci należą do grupy podwyższonego ryzyka. W najbliższej przyszłości można spodziewać się rozpowszechnienia terapii z wykorzystaniem inhibitorów immunologicznego punktu kontrolnego. Może to spowodować wzrost występowania cukrzycy jako działania niepożądanego takiej terapii. Tym samym wydaje się rozsądne zalecenie monitorowania stężenia glukozy we krwi oraz uczulenie pacjentów na ryzyko wystąpienia takiego działania niepożądanego.

Słowa kluczowe:

Wiad Lek 2018, 71, 5, -948



Localized on T cells and pro-B cells is the programmed cell death-1 (PD-1) immunoreceptor. Its blockade is a promising approach for a restoration of immune responses against cancer. One of these blocking antibodies is Nivolumab.

Nivolumab is a human IgG4 antibody. This class of antibodies is also referred as immune checkpoint inhibitors [1]. In many countries, Nivolumab is approved for the adjuvant treatment of patients with metastatic melanoma, advanced renal cell carcinoma, relapsed or progressive classical Hodgkin Lymphoma, metastatic non-small cell lung cancer, squamous cell cancer of the head and neck (SCCHN) and urothelial cancer [2].

Given the principle of action, autoimmune side effects can be expected as adverse events [3].

The main reported side effects in humans are immune-related adverse events, these include rash, vitiligo, thyroiditis, diarrhoea, hepatitis and pneumonitis (4). In 2015 Martin-Liberal et al. (5) described for the first time the case of an adult patient who developed autoimmune diabetes as a consequence of PD-1 inhibition with pembrolizumab.

In the same year, Huges et al. (6) reported the development of new-onset insulin-dependent diabetes in five patients after receiving anti-PD-1 antibodies. Four patients had received ipilimumab, one patient had received pembrolizumab. The time from drug administration to diabetes onset spanned 1 week to 5 months, when patients presented with severe hyperglycemia or diabetic ketoacidosis and with elevated HbA1c.

To date, the incidence of this autoimmune diabetes is still reported to be low [7].


To sensitize physicians for the risk of acute-onset diabetes mellitus presenting as an medical emergency under an increasingly frequent therapy with immune checkpoint inhibitors. This adverse event is reported to be rare yet, but the growing number of case reports raise doubts about the sheer rareness of the phenomenon. With an increasing therapy with anti-PD-1 antibodies, it might become an emerging challenge not only for the diabetologist.


A 65 yr old female, caucasian patient underwent surgical resection of the hepatic segment V and IV B, as well as cholecystectomy due to hepatocellular carcinoma in September 2012. Due to recurrency in the Douglas space, she underwent relaparotomy with R0 resection in the excavatio rectouterina and the septum rectovaginale in May 2015 and furthermore was treated with sorafenib from May 2015 until June 2016. Transcatheter arterial chemoembolization (TACE) was performed due to hepatic recurrency four times from June 2016 to November 2016. With again rising alpha fetoprotein (AFP) levels, the patient was treated with Nivolumab. Nivolumab is not approved for the treatment of HCC. However, as part of an individual healing trial, and after permission of the health care provider the patient was treated with nivolumab every 2 or 3 weeks since September 2017. The therapy was tolerated excellently and the computed tomographies of thorax and abdomen showed no evidence for tumorous tissue, AFP was within the normal range. On 30.5.2018 she had received nivolumab again. A few days later, she developed polyuria, headache, loss of vision, weight loss, paresthesia of the legs, fatigue and later on nausea and emesis. Upon admission in our emergency department, the patient could hardly be addressed; her blood glucose level was 23 mmol/l and she had ketonuria.

As for the further laboratory parameters, she had leucocytosis with 17.000 Gpt/l, a serum kreatinine of 103 µmol/l (normal Range 44-80) and normal values for sodium, potassium, calcium, CRP, bilirubin, ammonia and the serum transaminases.

The further laboratory workup revealed a Hba1c of 11,3% (100 mmol/molH [29-42]) and normal cortisone levels. Fasting C-peptide: 0,67 µg/l (1,1-4,4); AFP: 3,27 µg/l (<8,6). Autoantibodies: No antibodies against Glutamatdecarboxylase, Insulin, islet cells or tyrosine phosphatase-like protein (IA-2) autoantibodies. Furthermore, she had no antithyroid antibodies. HLA-Class Antigens: HLA-DR3DQB1*02:01 and DR4DQBI*03:02, suggestive of a predisposition to develop diabetes, as well as the rather protective DQB1*06:02 were negative.

In her medical record it is mentionable that she had no previous history of diabetes or relatives with diabetes mellitus. She underwent subtotal thyroidectomy due to multinodular goitre in 2007 and is under a stable supplementation regimen with 100 µg L-thyroxine.

In an attempt to rule out other possible factors for the onset of diabetes, such as metastatic growth in the pancreas, pancreatic endosonography was performed. It showed the intra-and extrahepatic bile ducts not extended. Furthermore, a normal sized pancreas was visible, the main pancreatic duct not dilated. In the corpus and tail we observed several hypoechoic inhomogeneities of unclear origin. For further diagnostic workup, we performed magnetic resonance tomography of the pancreas: Here, the pancreatic parenchyma was somewhat atrophic, appeared less dense and loose without evidence of focal lesions or necrosis-suspicious areas compared to previous imaging in 1/18.

Further laboratory and clinical investigations raised no suspicion for other side-effects of the nivolumab therapy. In the further course, the patient was treated with an intensified insulin therapy (ICT) regimen. She received insulin glargin and regular insulin preprandially, requiring about 44 IE insulin alltogether. The treatment regimen was unproblematic to handle by the patient and is continued to date.


The therapy with immune checkpoint inhibitors offers a cost-intensive, but also effective novel approach in the therapy of malignancies. This also applies for tumours where this therapy is not yet approved, as can be seen in this very case. However, side effects must be taken into account.

Our paper adds some new aspects to a growing body of reports about acute-onset diabetes with ketoacidosis under therapy with nivolumab [6−13]. The onset of autoimmune diabetes has been reported to be rare with less than 1% of the patients [7], but the increasing use of immune checkpoint inhibitors in clinical practice can be expected to cause also increasing numbers of patients with autoimmune diabetes as an adverse event. Although the pathomechanisms are different, the clinical manifestation of immune checkpoint inhibitor-induced diabetes is comparable with the “classical” type 1 diabetes with rapid onset and ketoacidosis and thus, a situation that requires immediate medical therapy. Autoimmune diabetes as an adverse event has not only been reported during therapy with nivolumab, but also for pembrolizumab and atezolizumab [6, 8]

The detection of diabetes-related autoantibodies is reported in some case reports, whereas other authors did not report such a detection. There is no relation between the time of onset of diabetes and the presence of autoantibodies [8]. For further assessment of the susceptibility to develop a type 1 diabetic situation, some authors had HLA-typing performed. Several HLA-genotypes are associated with a higher risk to develop type 1 diabetes mellitus, however, HLA-DR3-DQ2 and HLA-DR4-DQ8 occur in about 90% of the patients with type 1 diabetes mellitus [14]. These were not detected in our very case and yet, the findings from other case reports are inconclusive [8].

We also checked the literature for possible sex differences in the incidence of diabetes under immune checkpoint inhibitors with special regard to nivolumab and could not detect differences as were also not reported by Cheema et al. [13] for pembrolizumab. The rational to do this was because of the observation that males seem to benefit more from the therapy with checkpoint inhibitors: In a meta-analysis of 11 351 patients with advanced or metastatic cancers treated with immune checkpoint inhibitors, the authors reported the difference in efficacy between men and women treated with immune checkpoint inhibitors as significant (p=0·0019) [15]. However, the case numbers for the side effect “diabetes” are yet too small to allow statistical analysis.

As for the morphological aspect of the pancreas, there are no previous published reports about pancreatic imaging in nivolumab-induced diabetes, thus, the significance of our finding needs further observations.

Another interesting finding is the difference between onset of nivolumab therapy and diabetes, that ranges from one week up to 12 months. Our case with 8 months from onset of the therapy and the manifestation of insulin-dependent diabetes belongs to the late manifested diabetes cases. The reasons for this wide time span are not clear yet. Given the paucity of case reports, the most patients seem to develop the autoimmune diabetes 2-3 months after onset of treatment. Due to the nature of autoimmune diabetes, the only therapeutical approach is insulin therapy and all reported patients remained insulin-dependent after the initial presentation [8].

It has been speculated that due to increasing TNF-α, IL-2 and IFN-γ production under immunotherapy, this treatment could furthermore affect pancreatic beta cell function and insulin sensitivity [11]. This could also establish a rather type 2 diabetes prone situation [11]. This prompted the authors to perform a retrospective observational analysis from patients under immunotherapy registered in the French Pharmacovigilance Database (FPVD). The mean follow-up was 7 months. Among the 132 patients included, 3 cases of type 1 diabetes occurred. Interestingly, 6 new cases of prediabetes occurred in the course of treatment follow-up among 58 subjects who were tested for HbA1c. One patient was newly diagnosed with type 2 diabetes and was treated with metformin. Given the short observational period, the issue of type 2 diabetes warrants further investigation, too.


The therapy with immune checkpoint inhibitors is a promising approach in the therapy of several malignancies. An increasing use of these antibodies can be expected. However, these therapies can have considerable side-effects. Although still reported as rare, a rapid-onset and insulin-dependent diabetes mellitus is among these potential adverse events. It requires immediate therapy. Regular measurements of serum glucose are eligible and the patients should be informed about symptoms of the disorder. Physicians need to be aware about this potential side-effect of a therapy with checkpoint inhibitors.


1. Adachi  K, Tamada  K. Immune checkpoint blockade opens an avenue of cancer immunotherapy with a potent clinical efficacy. Cancer Sci. 2015;106:945-950.

2. Hassel JC, Heinzerling L, Aberle J et al. Combined immune checkpoint blockade (anti-PD-1/anti-CTLA-4): Evaluation and management of adverse drug reactions. Cancer Treat Rev. 2017; 57:36-49. doi: 10.1016/j.ctrv.2017.05.003.

3. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev. Cancer. 2012;12:252-264.

4. Brahmer JR, Tykodi SS, Chow LQ et al.. Safety and activity of anti-PD-L1 antibody in patients with advanced cancer. N Engl J Med 2012; 366: 2455–2465.

5. Martin-Liberal J, Furness AJ, Joshi K, Peggs KS, Quezada SA, Larkin J. Anti-programmed cell death-1 therapy and insulin-dependent diabetes: a case report. Cancer Immunol Immunother. 2015; 64:765-767. doi: 10.1007/s00262-015-1689-1

6. Hughes J, Vudattu N, Sznol M, Gettinger S, Kluger H, Lupsa B, Herold KC. Precipitation of autoimmune diabetes with anti-PD-1 immunotherapy. Diabetes Care. 2015;38:e55-7. doi: 10.2337/dc14-2349.

7. Godwin JL, Jaggi S, Sirisena I, Sharda P, Rao AD, Mehra R, Veloski C. Nivolumab-induced autoimmune diabetes mellitus presenting as diabetic ketoacidosis in a patient with metastatic lung cancer. J Immunother Cancer. 2017;5:40. doi: 10.1186/s40425-017-0245-2. eCollection 2017

8. Kapke J, Shaheen Z, Kilari D, Knudson P, Wong S. Immune Checkpoint Inhibitor-Associated Type 1 Diabetes Mellitus: Case Series, Review of the Literature, and Optimal Management. Case Rep Oncol. 2017;10: 897-909. doi: 10.1159/000480634. eCollection 2017 Sep-Dec.

9. Kumagai R, Muramatsu A, Nakajima R et al.Acute-onset type 1 diabetes mellitus caused by nivolumab in a patient with advanced pulmonary adenocarcinoma. J Diabetes Investig. 2017;8:798-799. doi: 10.1111/jdi.12627.

10. Miyoshi Y, Ogawa O, Oyama Y. Nivolumab, an Anti-Programmed Cell Death-1 Antibody, Induces Fulminant Type 1 Diabetes. Tohoku J Exp Med. 2016; 239: 155-158. doi: 10.1620/tjem.239.155.

11. Gauci ML, Laly P, Vidal-Trecan T et al.Autoimmune diabetes induced by PD-1 inhibitor-retrospective analysis and pathogenesis: a case report and literature review. Cancer Immunol Immunother. 2017;66:1399-1410. doi: 10.1007/s00262-017-2033-8.

12. Gaudy  C,  Clévy  C,  Monestier  S et al. Anti-PD1 pembrolizumab can induce exceptional fulminant type 1 diabetes. Diabetes Care 2015;38:e182-e183.

13. Cheema A, Makadia B, Karwadia T, Bajwa R, Hossain M. Autoimmune Diabetes Associated With Pembrolizumab: A Review of Published Case Reports. World J Oncol. 2018;9:1-4. doi: 10.14740/wjon1085w.

14. Nguyen C, MD, LC, Morahan G. Definition of High-Risk Type 1 Diabetes HLA-DR and HLA-DQ Types Using Only Three Single Nucleotide Polymorphisms. Diabetes. 2013;62:2135-2140.

15. Conforti F, Pala L, Bagnardi V et al.Cancer immunotherapy efficacy and patients‘ sex: a systematic review and meta-analysis. Lancet Oncol. 2018;19:737-746. doi: 10.1016/S1470-2045(18)30261-4.

Conflict of interest

All authors state that there is no conflict of intrest.

Corresponding Author

Igor Alexander Harsch

Thuringia Clinic Saalfeld “Georgius Agricola”

Rainweg 68, D-07318 Saalfeld/Saale

tel.:+49[0]3671/541569; Fax.:+49[0]3671/541403


Received: 16.07.2018

Accepted: 26.07.2018


Fig. 1. Pancreatic parenchyma somewhat atrophic, less dense and loose without evidence of focal lesions or necrosis-suspicious areas compared to previous imaging in 1/18.