Different patterns of response are associated with the dosages of immunosuppressants used to manage the adverse effects of immunotherapies.
Michael A. Postow, MD
Different patterns of response are associated with the dosages of immunosuppressants used to manage the adverse effects of immunotherapies, said Michael A. Postow, MD, at the International Congress on Immunotherapies in Cancer™.1
Postow, a medical oncologist with Memorial Sloan Kettering (MSK) Cancer Center in New York, New York, discussed interpretive strategies and solutions for toxicity problems that commonly occur with immunotherapies. High dose (HD) and low dose (LD) steroids may may not yield inferior outcomes to avoiding use of steroids; however, there is a trend toward better outcomes based on using smaller doses of steroids rather than larger ones, Postow said.
A study of immune-related adverse events (irAEs) in patients with melanoma treated with ipilimumab (Yervoy) at MSK found that patients treated with steroids for management of irAEs had equivalent or slightly improved time to treatment failure (TTF) versus those who did not receive immunosuppressants, Postow said. Of 298 patients, 103 (35%) required corticosteroid treatment for an irAE, the majority of them (n = 78) for grade ≥3 irAEs. The major reasons for corticosteroids were diarrhea (n = 50), hepatitis (n = 22), dermatitis (n = 21), and endocrinopathies (n = 14).
Median overall survival (OS) for the full study population was 16.5 months (95% CI, 12.6-21.1) and the estimated median TTF was 5.7 months (95% CI, 5.1-6.4). In an assessment of the effect of irAEs on OS and TTF, investigators found no difference in either measure when patients were stratified by the presence or absence of irAEs of any grade or by the administration of systemic corticosteroids to treat an irAE.2
Postow also noted a retrospective analysis of nivolumab (Opdivo) monotherapy in patients with advanced melanoma, designed to assess the safety profile of the agent and the management of irAEs. This found that use of immune modulators did not result in significantly different overall response rates: 29.8% (95% CI, 21.6%-39.1%) for those receiving corticosteroids (n = 114) versus 31.8% (95% CI, 27.6%-36.3%) for those who did not (n = 462).3
However, he said, a retrospective study evaluating HD steroids versus LD steroids in the treatment of ipilimumab-induced hypophysitis did find distinct survival patterns relative to the levels of dosage used. Among 98 patients with melanoma with ipilimumab-induced hypophysitis, OS and TTF were significantly longer in the LD group compared with the HD group. Median OS and TTF were not reached in the LD group and were 23.3 and 11.4 months, respectively, in the HD group. HD was identified as any dose larger than 7.5 mg daily of prednisone (Table).4
Investigators also found that all patients who had hypophysitis had superior OS compared with those who did not have hypophysitis (median OS, 28.2 vs 9.5 months; P = .0003). In addition, radiologic and endocrinologic outcomes and symptom resolution were not different between the LD and HD groups.4
The study was a helpful start in addressing the question of whether patients would do better if you didn’t treat their irAEs with immunosuppressants, a question that’s inherently difficult to resolve because patients often receive immunosuppression for adverse events, Postow said.
In this study, investigators found that it was relatively easy to distinguish between patients receiving LD and HD corticosteroids because for those with hypophysitis, the choice typically comes down to whether to treat with HD steroids to calm a pituitary inflammation quickly or to replace lost steroid with LD prednisone. “They found that hypophysitis itself was associated with favorable outcomes to ipilimumab and they found that if you give high doses of steroids for hypophysitis, those patients did worse than those patients who had low doses of replacement steroids,” he said. “This study raises the possibility that maybe you could get away with a little less steroid and maybe some patients would do better if that’s the case.”A related issue is recognizing the signs that hypophysitis is developing in the patient. When it occurs in the form of inflammation of the pituitary gland, it may often be overlooked because the adverse events as described by the patient (ie, headache, fatigue, depression) may be attributed to chemotherapy-related treatment for metastatic cancer. “It’s important to think about endocrinopathy, because many times, depression ends up being hypophysitis and people feel a lot better when you treat them with replacement steroids,” Postow said. It’s important to check a patient’s bloodwork and hormonal activity, but brain magnetic resonance imaging is also important. In such an event, it’s important to tell the radiologist to report on the size of the pituitary gland, enlargement of which is a sign of hypophysitis. Otherwise the radiologist may assume that’s not what you’re looking for and provide something else, he said.
Also affected by immunotherapy is the thyroid gland. Levels of thyroid stimulating hormone (TSH) in patients can be misleading, in that a case of acute thyroiditis will present with TSH that initially goes down before spiking well above normal when the patient’s thyroid function decreases. “It’s a factory on fire,” Postow said. “In those settings of hyperthyroidism, many physicians will wait for the thyroid to burn out and then give a thyroid replacement hormone. Some will give short courses of steroids if the patients are very symptomatic, or beta blockers if patients are having tachyarrhythmia.” High free T4 and low TSH levels in the initial phase of hyperthyroidism usually doesn’t mean it’s necessary to give an antithyroid drug such as methimazole, because the problem is more likely inflammation rather than Graves disease or something similar, he said. To provide supporting data, Postow presented findings of a study by Osorio et al on antibody-mediated thyroid dysfunction during T-cell checkpoint blockage in patients with non—small cell lung cancer.5 Six of 10 patients who received pembrolizumab (Keytruda) had transient hyperthyroidism before the onset of hypothyroidism, and no persistent hyperthyroidism occurred. The median onset of transient hyperthyroidism was 32 days (range, 21-59) among the 6 patients, and the median onset of hypothyroidism was 98 days (range 20-231).
Pituitary gland inflammation is more commonly seen in patients who are treated with CTLA-4 blockers, and the theory is that the pituitary has a mechanistic response to CTLA-4. The thyroid, however, behaves differently, and immune cells are not drawn to the thyroid as they are to the pituitary, Postow said. “One wonders what is it about the thyroid that is causing problems if it may not be this recruitment of inflammatory cells based upon expression of CTLA-4,” he said. “The thyroid gland [involvement] might be more of an auto-antibody— driven process.”
In the Osorio study, patients with thyroid dysfunction had antibody accumulations of the type associated with hyperthyroidism as well as with cases of autoimmune thyroiditis. That raises the question of whether, in this context, B-cell mediated immunosuppression strategies or T-cell mediated strategies could be used to get rid of these antibodies in patients with certain toxicities, he said. “It’s important to tease these apart so that we can think about mechanistic treatment of toxicities,” Postow said.
In the study, antithyroid antibodies developed in 8 of 10 patients with thyroid dysfunction and in 3 of 38 who did not have thyroid dysfunction (80% vs 8%; P <.0001). In 6 of 7 patients in whom antithyroid antibodies developed, antibody onset happened at the same time as the arrival of transient hyperthyroidism and preceded hypothyroidism.
Postow noted that chemotherapy-related toxicities are much better understood than irAEs. “Unfortunately, not a lot of prospective clinical trials have tested adverse event management strategies,” he said. “We can’t really say that there has been a trial and this is the best way to treat an adverse event, but there have been guidelines from the European Society for Medical Oncology, the American Society of Clinical Oncology, and the Society for Immunotherapy of Cancer. I don’t recommend one over another.” He also mentioned National Comprehensive Cancer Network guidelines for management of toxicities and the uptodate. com website.6-9
In summary, Postow said, irAE management is likely to become more complicated. “Increasingly, we’ll see rarer and rarer toxicities that develop, and we’re trying to learn as a community how to manage these adverse effects.”