Marked improvement of oral intake with nivolumab monotherapy in a patient with microsatellite instability‐high gastric cancer with insufficient oral intake

Abstract Although immune checkpoint inhibitors are commonly less effective for patients with a poor general condition, they can be effective and should be considered for poor general conditions in the case of MSI‐H tumor.


| INTRODUCTION
We report herein the case of marked improvement of oral intake via third-line nivolumab monotherapy in a gastric cancer patient with insufficient oral intake and European Cooperative Oncology Group performance status 2. We should consider immune checkpoint inhibitors for microsatellite instability-high gastric cancer, regardless of the patient's general condition.
Gastric cancer has one of the worst prognoses among all cancer types. Gastrointestinal obstruction due to primary obstruction or peritoneal metastasis is a common complication of advanced gastric cancer. Patients with gastrointestinal obstruction often experience insufficient oral intake, and these patients tend to have worse prognosis. In the JCOG0106 study, 1 sufficient or insufficient oral intake was defined based on whether drip infusion for nutrition support was performed, and the improvement of oral intake was defined as drip infusion not being indicated for > 7 days in patients who previously had insufficient oral intake. Some studies have shown improvement of oral intake with systemic chemotherapy, but these studies were in patients receiving firstor second-line treatment. [1][2][3][4][5][6][7] There have been no published cases of improvement of oral intake with third-or later-line chemotherapy.
Immune checkpoint inhibitors (ICIs) have become widely used for the treatment of various cancers. Nivolumab monotherapy for heavily pretreated gastric cancer was approved as a third-or later-line treatment in Japan in September 2017. However, according to the ATTRACTION-2 study, the response was very limited. 8 Further, although there are some reports of good response to ICIs in gastric cancer patients, there is no report on the improvement of oral intake by nivolumab monotherapy. 9 To our best knowledge, the current case is the first report of marked improvement of oral intake by third-line nivolumab monotherapy in a gastric cancer patient with insufficient oral intake.

| CASE HISTORY
A 76-year-old woman presented to our hospital with a stomachache and loss of appetite for 3 months. She was admitted with insufficient oral intake for careful examination and treatment. She had rheumatoid arthritis treated with methotrexate.

| Differential diagnosis, investigations, and treatment
Physical examination revealed no significant findings. The laboratory data on admission are summarized in Table 1. Computed tomography (CT) revealed gastric wall thickening, invasion of the abdominal wall, and regional lymphadenopathy ( Figure 1A). Gastrointestinal endoscopy (GIE) revealed Bormann type 3 gastric cancer ( Figure 2A). Biopsy of the gastric mucosa showed poorly differentiated adenocarcinoma ( Figure 3A, 3B). Immunohistochemical analysis (HercepTest, Dako, Glostrup, Denmark) showed no expression of human epidermal growth factor receptor 2 (score = 0), and HER2/neu amplification was confirmed negative using dual color in situ hybridization (INFORM HER2 Dual ISH DNA Probe Cocktail Assay). 10 The Epstein-Barr encoding region in situ hybridization was also negative ( Figure 3C). 11 Programmed death-ligand 1 (PD-L1) protein expression on adenocarcinoma cells was assessed using PD-L1 IHC 22C3 pharmDx (Agilent Technologies; Carpinteria, CA, USA), and the combined positive score (CPS) was 10 ( Figure 3D). 12 Microsatellite instability was high (MSI-IVD Kit, FALCO Ⓡ ). Therefore, she was diagnosed with unresectable advanced microsatellite instability-high (MSI-H) gastric cancer. 12 She underwent gastrojejunal bypass surgery in July 2017, but her oral intake did not sufficiently improve. Therefore, total parenteral nutrition (TPN) was started. A month after the bypass, mFOLFOX6 therapy (oxaliplatin 85 mg/m 2 intravenous [IV], l-leucovorin 200 mg/m 2 IV, and 5-fluorouracil 400 mg/m 2 IV bolus followed by 2400 mg/m 2 over 46 hours starting on Day 1, every 2 weeks) was started as first-line treatment. However, CT revealed disease progression after six cycles. Paclitaxel therapy (paclitaxel 80 mg/ m 2 on days 1, 8, and 15 every month) as second-line treatment was also ineffective at the first evaluation after 3 cycles. Her general condition did not improve, and she needed TPN after progression on paclitaxel therapy. Although her Eastern Cooperative Oncology Group (ECOG) performance status (PS) was 2, nivolumab monotherapy (3 mg/m 2 every 2 weeks until October 2018 and 240 mg/body every 2 weeks from November 2018) was administered as third-line treatment. Laboratory tests, CT, and GIE before the first administration of nivolumab are summarized in Table 2, Figure 1B

| Outcome and follow-up
After four cycles of nivolumab, CT showed that the tumor had shrunk significantly. Her ECOG PS improved from 2 to 0. After 6 months, her oral intake problem had completely improved, and TPN was no longer needed. Upper gastrointestinal endoscopy indicated tumor shrinkage ( Figure 2C). CT and upper gastrointestinal endoscopy revealed no residual tumor 2 years after the surgery (Figures 1C and 2D). These findings support that nivolumab monotherapy caused a complete response. She is currently receiving nivolumab monotherapy and has sufficient oral intake ( Figure 4). Facial paralysis occurred 7 months after nivolumab administration, and we diagnosed it as Bell's paralysis not related to nivolumab. She developed grade 1 liver dysfunction and grade 1 diarrhea as immune-related adverse events (irAEs).

| DISCUSSION
We report herein a case of marked improvement of oral intake by third-line nivolumab monotherapy in a patient with MSI-H gastric cancer with insufficient oral intake. The first-and second-line treatments were not effective, but she achieved a complete improvement of oral intake with nivolumab and had a complete tumor response after nivolumab monotherapy. To our best knowledge, this is the first case report on improved oral intake from third-line nivolumab therapy.
Insufficient oral intake is one of the most common complications of gastric cancer. The main causes of insufficient oral intake are gastric outlet obstruction and peritoneal metastasis. Insufficient oral intake is a poor prognostic factor, and Shitara et al reported that the median overall survival (OS) was significantly shorter in patients with insufficient oral intake than in those with sufficient oral intake (5.0 months vs. 12.7 months, P < .05). 3 In previous studies, the rate of improvement of oral intake with first-line treatment was 32%-85% (Table 3). [1][2][3][4][5][6][7] Platinum-containing regimens seem to be more effective than platinum-free regimens for the improvement of oral intake (Table 3). However, there are no reports of improvement of oral intake in patients receiving third-or later-line treatment. In this case, oral intake was not improved by first-and second-line treatment; additionally, the patient experienced disease progression. However, oral intake was completely improved with third-line nivolumab monotherapy, and complete tumor response was achieved as observed on imaging studies.
Nivolumab monotherapy is administered for heavily treated gastric cancer according to the ATTRACTION-2 study. However, in that study, the response rate was 11%, and only three patients had a complete response, 8,13 although patients with complete or partial response had long-term response. 13 Unfortunately, few patients responded, and hyperprogressive disease was common. 14 Many studies have examined predictive factors of the response to ICIs to select patients who may respond to such treatment. Microsatellite instability is one such predictive factor. Janjigian et al reported that the efficacy of cytotoxic drugs was limited among patients with MSI-H gastric cancer. 15 According to the KEYNOTE-158 study, which was a single-arm phase II trial for MSI-H solid tumors treated with pembrolizumab, the median OS of all patients was 27.8 months, and the median OS of MSI-H gastric cancer patients was not reached. 16 In the subanalysis of the KEYNOTE-061 and −062 studies, which were phase III trials of second-line and first-line pembrolizumab, respectively, for gastric cancer patients, the median OS of patients with MSI-H gastric cancer who received pembrolizumab was longer than that of those receiving chemotherapy. 12,17 In the current case, the patient had a durable response to thirdline nivolumab because she had MSI-H gastric cancer. The first-and second-line treatments were not effective, which is consistent with the previous report. The patient has been disease free for 2 years with nivolumab therapy. Therefore, we believe it is important to evaluate the microsatellite instability status before first-line treatment, and if the patient has an MSI-H tumor, the response should be evaluated early during first-line treatment. PD-L1 and Epstein-Barr virus infection have also been reported to be predictive factors for treatment response to ICIs. 18 In the KEYNOTE-062 study, 21 (11.5%) of 182 patients with a CPS ≥ 10 had MSI-H tumors, whereas 33 (6.5%) of 506 patients with a CPS ≥ 1 had MSI-H tumors. 17 CPS is associated with MSI status, and gastric cancer patients with a CPS ≥ 10 have sustained response to ICIs. T A B L E 2 Laboratory data at the first administration of nivolumab