乌鸦传媒

For all the tremendous benefits immunotherapy offers patients with non-small cell lung cancer (NSCLC), immune-related adverse events (irAEs) can be equally burdensome. Treatment with immune checkpoint inhibitors (ICIs) targeting PD-1/PD-L1 can take a particular toll on a patient's endocrine system, potentially leading to insulin-dependent diabetes.

"Endocrine irAEs, such as hypophysitis, thyroid dysfunction, and insulin-dependent diabetes mellitus [DM], can present with unique profiles that are not seen with the use of traditional chemotherapeutics," noted Monica Girotra, MD, of Memorial Sloan Kettering Cancer Center in New York City, and colleagues in .

While new-onset autoimmune DM as an irAE is rare (), the appearance of diabetic ketoacidosis (DKA) is a medical emergency requiring immediate treatment, emphasized James Luke Godwin, MD, of Fox Chase Cancer Center in Philadelphia, and colleagues in a systematic review and meta-analysis in JAMA Oncology.

In another paper, they presented a case study of a woman with metastatic NSCLC who was treated with second-line nivolumab (Opdivo) after disease progression following standard chemoradiotherapy. Two ICI doses in, the 34-year-old woman developed abrupt onset of hyperglycemia and DKA.

She was given a diagnosis of autoimmune diabetes and required continuous subcutaneous insulin infusion based on "erratic glycemic excursions and multiple readmissions for DKA." While the NSCLC was in remission at the time of the report, and the patient did not require additional cancer treatment, she continued to "have erratic blood sugars with frequent hypoglycemia and hyperglycemia," the authors stated in the .

In another case, from Greece, a patient with NSCLC and no prior history of diabetes underwent 10 months of treatment with nivolumab before presenting with polyuria and polydipsia; a diagnosis of ICI-related type 1 diabetes (T1D) was made. The patient was able to come back to ICI therapy, but "remained insulin-dependent until his death from complications of his disease," noted Anastasios Boutis, MD, PhD, of Theagenio Hospital in Thessaloniki, and co-authors in their report in .

Caicun Zhou, MD, PhD, of Shanghai Pulmonary Hospital in China, and colleagues reported on a male patient who also had DKA after 10 cycles of ICIs, and the irAE of DM was made. The patient continued with cancer treatment once the DKA resolved, and his blood glucose was controlled, the team stated in .

In , Ho-Cheol Kang, MD, PhD, of Chonnam National University Medical School in Gwangju, South Korea, and colleagues reported on four cases of ICI-related DKA, mostly after treatment with pembrolizumab (Keytruda), including a lung cancer patient who developed DKA after three cycles of ICI. The authors reported an observed incidence of DKA of less than 1% in their study -- i.e., similar to the overall rate -- "but the mean duration of DKA onset after the start of ICIs was 15.8 weeks, which was relatively longer than previously reported," the researchers wrote.

Patients who start ICI therapy with diabetes that is well controlled can still face crisis-level irAEs. Yogish C. Kudva, MD, of the Mayo Clinic in Rochester, Minnesota, and colleagues reviewed electronic health records data from 1,444 patients, and pinpointed a dozen patients who received ICIs over 6 years who developed new-onset insulin-dependent diabetes; nine others experienced worsening of preexisting type 2 diabetes (T2D). In the latter group, the "underlying mechanism appears similar to spontaneous T1D but there is a faster progression to severe insulin deficiency," the authors wrote in .

In another study, Sarita Goud, MD, of Mercy Catholic Medical Center in Darby, Pennsylvania, and colleague, reported on a lung cancer patient with well-controlled T2D on metformin who presented with frequent DKA 2 weeks after treatment with cycle three of pembrolizumab. Her worsening glycemic control led to insulin dependency, the team wrote in the abstract presented at the .

Finally, at the 2021 virtual American Society of Clinical Oncology (ASCO) annual meeting, a study by , of Princess Margaret Cancer Centre in Toronto, and colleagues focused on the characteristics of patients treated at five Canadian cancer centers who developed ICI-related insulin-dependent diabetes. Among 27 patients (four with lung cancer), seven had preexisting diabetes or prediabetes, and 13 presented with DKA after treatment started.

Most of the patients received single-agent immunotherapy, and had a median time to onset of irAE-driven diabetes of 4.8 months. Patients who were on a combination with anti-PD-1/L1 therapy had a median time to onset of 1.4 months. "Other than insulin replacement, four patients received steroids," Pimentel Muniz noted in the presentation. "And two patients received infliximab [Remicade] in an attempt to reverse" the ICI-induced diabetes. However by study's end, "all patients were long-term insulin dependent."

Awareness and Counseling

The various authors offered several take-home messages from their research:

• Godwin and co-authors: Their female patient had three diabetes-related autoantibodies before she even began ICI therapy, but "half of the reported cases of autoimmune DM after anti-PD-1 therapy occurred in patients with no detectable diabetes-related autoantibodies ... Medical oncologists must be aware of the possibility of anti-PD-1 therapy induced autoimmune diabetes and counsel their patients to report symptoms that may be related to DKA."

• Goud and co-authors: "Physicians should counsel patients about this potential immune-related adverse effect and educate them about the symptoms of hyperglycemia and DKA."

• Boutis and co-authors: While glucose control can be achieved with insulin, clinicians need to be aware that "long-term exogenous insulin-dependence is inevitable even after discontinuation of treatment. It seems that diabetes mellitus results in complete beta-cell destruction and is therefore not reversible." The "exact pathophysiologic mechanism and predictive biomarkers" for ICI-related T1D have yet to be established. The "result is permanent insulin-dependence and, in a departure from treatment for other immune-related adverse events, corticosteroids are not recommended."

• Zhou and co-authors: "Glycemic monitoring is imperative during ICI treatment," and "after controlling the blood glucose level, continuing the immune therapy could still be of benefit and safe for the patient."

• Kang and co-authors: The time to the T1D diagnosis post-ICI therapy was not tied to the level of HbA1c, implying "the acute onset of hyperglycemia."

• Kudva and co-authors: In the patients with existing T2D, 78% developed another irAE, most commonly thyroiditis.

• Pimentel Muniz and co-authors: Their results did not support the use of steroids and infliximab in these patients, and "combination immune checkpoint regimens accelerate the development of [ICI-related diabetes] in predisposed individuals."

Risk Assessment

Islam Elkonaissi, lead pharmacist at the Royal Marsden Hospital in London, and colleagues wanted to get a better picture of how these patients were managed at their institution. "The question for us -- in light of an incident where we had a patient develop diabetes as a result of immunotherapy toxicity -- was: 'Are we monitoring patients and taking risk assessment as per national, local guidelines?'" he said in a video for .

"One of the guidelines that we've used is the Joint British Diabetes Societies [JBDS], and these advise that risk assessment should be carried out ... with the aim to provide early and timely intervention ... That intervention could be referring the patent to the endocrinologist or advising the patient on dietary support, or adjusting or initiating hypoglycemic agents, or simply ... letting the regular diabetes team of that patient know about it," Elkonaissi continued.

Patients with NSCLC, who are often older, are at a very high risk of multiple comorbidities, not to mention treatment toxicities, "so should we just consider everyone as high risk and manage them accordingly? Or should we undertake risk assessment -- are we going to waste our time by undertaking risk assessment, when we could just monitor them and manage them ... this was the subquestion of our analysis," he explained.

Using electronic health records data from 125 patients with NSCLC treated with ICIs, Elkonaissi's group found that 29% of the patients had received steroids to treat grade 1 hyperglycemia but no risk assessment was done using the JBDS assessment criteria, the researchers noted in their presentation at the 2019 International Society of Oncology Pharmacy Practitioners symposium.

While Elkonaissi acknowledged that the management of NSCLC, including treatment algorithms, has gotten more complex over time, he stressed the importance of using guidelines, such as those from JBDS, the National Comprehensive Cancer Network, and ASCO, because "it is very hard to grade patients' toxicity and level of clinical signs of hyperglycemia without considering the overall picture."

Comment