Gastritis as an immunotherapy-related toxicity in the treatment of endometrial cancer: A case report

Highlights

• •Gastritis as a rare but serious side effect of immune checkpoint inhibitor therapy for which no formal grading exists.
• •Nausea, vomiting, dysphagia, early satiety and weight loss are signs of gastritis.
• •Need for biopsy to confirm the diagnosis highlights the importance of a multidisciplinary team approach.

Abstract

Gastritis related to immunotherapy use is a less commonly reported adverse effect. With increasing use of immunotherapy agents in the management of patients with endometrial cancer, even rare adverse effects are being seen more frequently in gynecologic oncology practice. A 66-year-old with recurrent mismatch repair deficient endometrial cancer was treated with single-agent pembrolizumab. She initially appeared to tolerate treatment well; however after 16 months of therapy she began to develop nausea, vomiting, and abdominal pain that resulted in 30-pound weight loss. Pembrolizumab was held out of concern for immunotherapy related toxicity. She underwent evaluation with gastroenterology including esophagogastroduodenoscopy (EGD) with biopsy that demonstrated severe lymphocytic gastritis. She was treated with IV methylprednisolone with improvement in symptoms over three days. She was then transitioned to oral prednisone at 60 mg daily with weekly taper by 10 mg, with a proton pump inhibitor (PPI) and carafate until resolution of symptoms. She subsequently had a follow up EGD with biopsy, which demonstrated resolving gastritis. She is presently doing well off of steroids with stable disease noted on her last scan after cessation of pembrolizumab.

1. Introduction

The use of immune checkpoint inhibitors (ICIs) such as pembrolizumab (a programmed cell death 1 (PD-1) inhibitor) has been shown to be effective in the treatment of several gynecologic malignancies, and is FDA approved for the treatment of progressive endometrial cancer which has been shown to be microsatellite instability-high (MSI-H) or dMMR (Gaffuri, 2019). By inhibiting the downregulation of T-cell-mediated destruction, these pharmacotherapeutics enable the patient’s immune system to target and destroy neoplastic cells (Fujii, 2020). This very mechanism, however, can also result in a multitude of autoimmune side effects as well. As an immune system modulator, immunotherapies have the potential to disrupt other organ systems aside from their original target, including the gastrointestinal system. Side effects affecting the distal gastrointestinal (GI) tract including colitis and diarrhea are well recognized (Kumar, 2017), however gastritis induced from immune checkpoint inhibitors has also been described (Pardoll, 2012, Woodford, et al., 2021, Weinmann and Pisetsky, 2019, Makker, 2022). Gastritis is a broad category of diagnoses that stem from inflammation to the gastric mucosa that can lead to nausea, vomiting, abdominal pain, and weight loss. Here we describe a case of the development of severe gastritis after the initiation of pembrolizumab in the treatment of recurrent endometrial cancer.

2. Case

A 66-year-old was diagnosed with FIGO stage IVB grade 3 endometrioid adenocarcinoma with a large vaginal mass extending to the introitus, tumor that had infiltrated the colon and ovaries, as well as tumor nodules on pelvic peritoneal surfaces and bowel epiploica. This was initially treated with hysterectomy, bilateral salpingo-oophorectomy and tumor resection followed by six cycles of carboplatin and paclitaxel. Imaging performed one month after completion of chemotherapy revealed increasing pelvic adenopathy and right vaginal apex mass concerning for progression. Tumor testing revealed mismatch repair deficiency (loss of MLH1 and PMS2) due to MLH1 promoter hypermethylation. She was started on pembrolizumab 200 mg IV every 3 weeks on 7/14/2020. After six cycles she was transitioned to 400 mg IV every 6 weeks on 11/17/2020. She continued treatment without major adverse effects for 25 cycles, until 1/11/2022 when cycles 26/27 were held. During this treatment period she had interval imaging which showed treatment response.

Approximately 16 months after initiating therapy with pembrolizumab, she presented with nausea, vomiting, dysphagia to solid foods, early satiety, and an unintentional 30-pound weight loss over the course of two months. Pembrolizumab was held and CT of the abdomen and pelvis was ordered. CT showed wall thickening of the stomach consistent with possible inflammation. Referral to gastroenterology was placed and she subsequently underwent an EGD, which demonstrated diffuse gastric erythema with adherent thick mucus, mucosal friability, thickened stomach folds and mild prepyloric nodularity, with normal duodenal mucosa. Final pathology of EGD biopsies showed severe acute lymphocytic gastritis, negative for H. pylori, dysplasia, or malignancy. A small bowel biopsy showed intact mucosa with normal architecture. The patient was started on IV methylprednisolone at 1 mg/kg, IV pantoprazole, and sucralfate until improvement in symptoms. After three days of IV treatment, she was able to tolerate oral food and medications and was discharged on oral prednisone 60 mg daily and sucralfate. After two weeks at 60 mg daily with no recurrence of symptoms, a taper was initiated with a decrease of 10 mg per week. Steroids were ultimately discontinued as she continued to feel well. She was recommended to continue a low dose PPI (40 mg pantoprazole PO daily) indefinitely. Follow up EGD was performed 6 months later which showed resolving of the previously noted gastritis on biopsies of the stomach antrum and body. A small polyp was biopsied on the gastric mucosa which was consistent with chronic inflammation.

Repeat CT abdomen/pelvis was performed 9/1/22 and revealed stable disease, after holding treatment with pembrolizumab for two months. Given stable disease, the decision was made for continued surveillance without re-starting pembrolizumab.

3. Discussion

This case demonstrates the rare adverse event of gastritis after the initiation of immunotherapy with pembrolizumab. The incidence of gastritis as a side effect of immunotherapy has been less described than other GI toxicities. Toxicities related to immuno-oncology (IO) have been more often described in patients who receive combination therapy of anti-CTLA4 and anti-PD1 therapy than anti-PD1 monotherapy (Kumar, 2017). Recommendations for grading toxicities have been described by the American Society of Clinical Oncology in order to better guide clinicians in recognizing and managing a wide variety of side effects from commonly used immunotherapies (Tang, 2021). With the increasing use of ICIs in the treatment of gynecologic malignancies, it is important to describe the breadth of adverse effects of these medications. Given the rarity of IO-gastritis, a formal grading system for this particular adverse effect does not exist (Tang, 2021). Because of this, the best approach to management is not yet known and clinicians may manage gastritis of similar severity with different approaches. Updated ASCO guidelines state median time to onset of GI toxicities is 6 weeks, but with a wide range (1–107.5 weeks) (Schneider, 2021, Rudzki, 2018).

Though a rare toxicity resulting from immunotherapy, several other case reports have emerged describing IO- gastritis. Woodford et al performed a literature review of 25 cases of IO-gastritis which found that this toxicity has been increasingly reported and can result in vague but devastating complications. They found that this toxicity was not more commonly associated with combination therapy as typically seen in other IO-related GI toxicity such as colitis. Variable timing of symptom onset was described, ranging from 2 weeks to 156 weeks. Further, a variety of symptoms were described including similar symptoms to those found in our patient such as nausea, vomiting, dysphagia, and abdominal pain (Weinmann and Pisetsky, 2019).

Furthermore, it is known that these toxicities are dose-dependent, with higher dosages resulting in more frequent or more severe toxicities (Tang, 2021). Our patient had transitioned from every 3 week dosing to every 6 week dosing prior to the development of IO-gastritis. It is unclear whether the higher dosing and less frequent schedule contributed to risk of toxicity.

The multi-disciplinary workup and treatment of the patient in this case report was consistent with those described in current literature regarding immunotherapy-related gastritis (Weinmann and Pisetsky, 2019). Patients underwent EGD and biopsies demonstrating pathology consistent with gastric inflammation similarly to the patient described here. Interestingly, a wide spectrum of severity of inflammation resulting from IO-gastritis has been described. Biopsies with chronic duodenitis or chronic gastritis with rare granulomas have been reported in association with IO-gastritis as well (Simonaggio, 2019).

In the literature describing this IO-toxicity, patients were treated IV steroid treatment along with proton pump inhibitor and sucralfate, which was similar to the patient described here (Weinmann and Pisetsky, 2019, Makker, 2022). The updated ASCO guideline does include recommendation for high dose PPI during any treatment with high-dose steroids (Tang, 2021) to prevent ulceration that can often be a challenge seen with steroid treatment. For this reason it has been demonstrated that biopsy-confirmed diagnosis is critical prior to starting steroid treatment for toxicity involving the GI tract (Qin, 2020). Woodford et al described that the reported dose of IV steroids and duration of treatment was variable among published cases. Most frequently, patients were treated with 1–2 mg/kg/day of IV solumedrol, followed by a prolonged PO prednisone taper, decided on by the individual clinician (Weinmann and Pisetsky, 2019). Recurrence of symptoms while on treatment was reported in few cases (13%), suggesting that once symptoms are identified and appropriate treatment started, recurrence risk is low.

Post-treatment EGD to evaluate for resolution of gastritis has been reported and may be helpful when symptoms are resolved and restarting immunotherapy is being considered. Among the previous case reports describing this, only a small percentage of those patients underwent a post-treatment EGD to determine resolution of the adverse effect. Despite this, very few of the total number of patients had any recurrence of symptoms following completion of steroid therapy, suggesting that a post-treatment EGD may not be entirely necessary (Weinmann and Pisetsky, 2019). In order to guide subsequent oncologic therapy however, it could be considered.

In the absence of a grading system or standardized guidelines for decisions regarding restarting treatment, the decision to restart following resolution of gastritis should be a shared decision with the patient, their oncologists and the GI team. Some studies have shown moderate to high rate of recurrence of IO-toxicity when rechallenged with the same ICI (Verschuren, 2016). On review of literature, several case reports have described success with rechallenging patients to oncologic immunotherapy after grade 3 or higher IO-toxicity (Verschuren, 2016). Much of this data described clinicians concomitantly prescribing low-medium dose (5–30 mg) oral steroids with their oncologic immunotherapy while rechallenging them in order to prevent recurrence of toxicity. While theoretically, this use of steroids due to their anti-inflammatory mechanism of action would counteract the effectiveness of immunotherapy in oncologic treatment, some studies have shown benefit (Onuki, 2018). This suggests that rather than interfering with the ability of tumor targeted immunoactivity by agents such as pembrolizumab, concomitant use of steroids beneficially work towards suppression of immune system overactivity. At this time there are no clinical trials to establish a concrete methodology for rechallenging patients who have experienced IO-toxicity and at this time is left to the clinician’s discretion.

After adverse effects of any IO are diagnosed, prior studies have suggested a risk of up to 50% of development of a second, different IO-toxicity (Nelson, et al., 2021). In the largest gastritis case series, 22% of patients were diagnosed with a second IO-related AE (Weinmann and Pisetsky, 2019). This case series included a heterogenous group of tumors. It is not well described whether the tumor type, indication for pembrolizumab use (MMRd vs PDL1 receptor positivity), or other patient factors may modify risk for development of IO toxicity. Given the rarity of these events, reporting rare toxicities will provide an important avenue for improving our management of the toxicities and subsequent oncologic therapy.

4. Conclusion

Various toxicities have been described with the use of immunotherapy in cancer treatment (Tang, 2021, Qin, 2020). IO-gastritis is a rarer toxicity that is being recognized more often with increased use of IO in treatment of endometrial cancer. This case report highlights the importance of maintaining a high index of suspicion for even rare IO-toxicities in treating patients with immunotherapy. Further, a multi-disciplinary approach involving specialists, like gastroenterology in this case, allows a more comprehensive assessment of the toxicity and appropriate diagnostics to help guide management. Whether there are tumor specific or patient specific risk factors for the development of IO gastritis is not yet known. Determining when to restart IO after resolution of gastritis and whether to concomitantly treat with steroids is currently made on a case-by-case basis. Whether oncologic outcomes are different when treatment is discontinued or restarted with low dose steroids is not yet well described. Given the improvement with initiation of steroids, prompt recognition and diagnosis may prevent longer than necessary oncologic treatment delays.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.