Advanced intestinal-type ampullary cancer successfully treated using combination therapy with gemcitabine,cisplatin,and durvalumab: A case report

Abstract

A 65-year-old Japanese man presented with right-sided abdominal pain. Findings from computed tomography, esophagogastroduodenoscopy, and endoscopic ultrasound-guided tissue acquisition led to a diagnosis of ampullary cancer with para-aortic lymph node metastasis. To treat the biliary obstruction, a metallic stent was placed endoscopically. Combination therapy comprising gemcitabine-, cisplatin-, and durvalumab-induced marked regression of the lymph node metastases, and the patient’s carbohydrate antigen 19-9 levels decreased to within normal limits. After eight gemcitabine, cisplatin, and durvalumab therapy cycles, the treatment was switched to maintenance durvalumab monotherapy. This monotherapy was continued for four cycles without adverse effects, and it maintained the metastatic lymph node’s regression and the decreased carbohydrate antigen 19-9 levels. No stent occlusion was observed. Histopathological examination of the biopsy specimen obtained before treatment initiation revealed intestinal-type ampullary cancer, and programmed cell death ligand 1 expression was positive in 10% of tumor cells and 5% of immune cells. Despite the limited evidence regarding the long-term efficacy of gemcitabine, cisplatin, and durvalumab therapy in ampullary cancer, this case report demonstrates its sustained effectiveness over 18 months in a patient with intestinal-type ampullary cancer.

Introduction

Ampullary cancer (AC) is a rare gastrointestinal malignancy originating from the papilla of Vater, accounting for only 0.2% of all gastrointestinal cancers. ¹ AC generally has a more favorable prognosis than other periampullary malignancies.^2,3 However, similar to other cancers, the prognosis worsens in the presence of distant metastases. ⁴ Furthermore, evidence regarding chemotherapy specifically for AC is limited. In Japan, AC is classified as a type of biliary tract cancer, and chemotherapy is typically administered based on the treatment regimens for biliary tract cancer. ⁵

Recently, the TOPAZ-1 trial demonstrated the effectiveness of combination therapy with gemcitabine, cisplatin, and durvalumab (GCD therapy). ⁶ Consequently, GCD therapy is increasingly being adopted as a first-line treatment for biliary tract cancer. However, no patients with AC were included in the TOPAZ-1 trial. Thus, the efficacy of GCD therapy in this malignancy remains unclear. In this study, we report the case of a patient with unresectable AC who maintained a partial response for 18 months during continuous GCD therapy.

Case

A 65-year-old Japanese man with no significant medical history presented to another hospital with right-sided abdominal pain. His laboratory findings on admission were as follows: white blood cell count, 16,900/μL; aspartate aminotransferase, 81 U/L; alanine aminotransferase, 199 U/L; gamma-glutamyl transpeptidase, 474 U/L; and carbohydrate antigen 19-9 (CA19-9), 184.5 U/m (Table 1). Noncontrast computed tomography (CT) revealed an enlarged gallbladder with wall thickening but no evidence of gallstones or common bile duct stones. The patient was diagnosed with cholecystitis, and percutaneous transhepatic gallbladder drainage (PTGBD) was performed. Although his cholecystitis was alleviated relatively quickly, his hepatobiliary enzyme levels gradually increased. Contrast-enhanced CT revealed mild common bile duct dilation, enlarged lymph nodes extending from the dorsal aspect of the pancreatic head to the region surrounding the superior mesenteric vein, and an enlarged para-aortic lymph node (Figure 1(a) and (b)). Esophagogastroduodenoscopy (EGD) revealed a tumor at the papilla of Vater (Figure 2(a)), and histopathological examination of biopsy samples confirmed adenocarcinoma with poorly to well-differentiated features. Additionally, endoscopic ultrasound-guided tissue acquisition of enlarged lymph nodes dorsal to the pancreatic head was performed (Trident Needle Biopsy System; Micro-Tech (Nanjing) Co. Ltd., Nanjing, China), and the histopathological findings were consistent with adenocarcinoma, similar to the ampullary lesion. Although endoscopic ultrasound-guided tissue acquisition (EUS-TA) targeting lymph node #16 was technically challenging; a comprehensive assessment of the diagnostic findings led to a diagnosis of para-aortic lymph node metastasis. The final diagnosis was unresectable advanced AC with multiple lymph node metastases, including para-aortic nodes. According to the union for international cancer control (UICC) TNM staging system (8th edition), the tumor was classified as cT2N1M1 (cStage IV). Subsequently, endoscopic retrograde cholangiopancreatography was performed. Endoscopic biliary drainage was achieved by placing an uncovered self-expanding metallic stent (SEMS; Niti-S biliary uncovered stent; Taewoong Medical Co. Ltd., Gimpo, South Korea), after which the PTGBD tube was removed. Detailed histopathological analysis of the ampullary tumor biopsy revealed features consistent with the intestinal type, including MUC1 negativity and MUC2 positivity (Figure 3(a)–(c)). Furthermore, programmed cell death ligand 1 (PD-L1) expression was detected in approximately 10% of tumor cells and 5% of immune cells (Figure 3(d)). GCD therapy comprising gemcitabine (1000 mg/m², days 1 and 8), cisplatin (25 mg/m², days 1 and 8), and durvalumab (1500 mg, day 1) in a 21-day cycle was initiated. During the third cycle of GCD therapy, the patient’s CA19-9 level had decreased to 11.5 U/mL. After the third cycle, the patient developed grade 4 neutropenia. Consequently, gemcitabine and cisplatin dosages were reduced, and the administration schedule was modified (gemcitabine: 500 mg/m², days 1 and 15; cisplatin: 12.5 mg/m², days 1 and 15; durvalumab: 1500 mg, days 1; one course for 28 days). After the fifth cycle, CT revealed significant lymph node metastasis regression. At the patient’s request, he was transferred to our hospital, where treatment was continued from the sixth cycle. The same dosing and schedule were maintained until the eighth cycle, after which maintenance therapy with durvalumab monotherapy (1500 mg every 28 days) was initiated. The patient was monitored with tumor marker assessments every month and CT scans every 2–3 months. After the fourth cycle of maintenance durvalumab monotherapy, follow-up EGD revealed no enlargement of the primary ampullary tumor, and effective biliary drainage was maintained with an uncovered SEMS (Figure 2(b)). After the ninth cycle of maintenance durvalumab monotherapy (approximately 18 months after initiating GCD treatment), CA19-9 and CEA levels remained within normal limits, and regression of the metastatic lymph node was maintained with no new metastatic lesions (Figure 4). Furthermore, the patient did not experience cholangitis or other notable adverse events during the treatment course. Maintenance durvalumab monotherapy is still ongoing.

Table 1.

Laboratory data on admission.

Parameter	Values	Parameter	Values
Blood cell counts		TBIL (mg/dL)	1.7
WBC (/μL)	16,900	CRP (mg/dL)	2.33
RBC (×104/μL)	453	LDH (IU/L)	201
Hb (g/dL)	14.6	Glu (mg/dL)	135
PLT (×104/μL)	25.2	ChE (IU/L)	310
		CK (U/L)	46
Coagulation function		AMY (U/L)	84
PT (s)	0.96	BUN (mg/dL)	18.0
PT (%)	115	CRE (mg/dL)	0.77
PT/INR	0.96	Na (mEq/L)	135
		K (mEq/L)	4.1
Biochemical parameters		Cl (mEq/L)	100
TP (g/dL)	7.1	Ca (mg/dL)	9.2
Alb (g/dL)	4.3
AST (IU/L)	81	Tumor markers
ALT (IU/L)	199	CEA (ng/mL)	4.3
ALP (IU/L)	135	CA19-9 (U/mL)	184.5
γ-GTP (IU/L)	474

WBC: white blood cell; RBC: red blood cell; Hb: hemoglobin; Plt: platelet; PT: prothrombin time; INR: international normalized ratio; TP: total protein; Alb: albumin; AST: aspartate aminotransferase; ALT: alanine aminotransferase; ALP: alkaline phosphatase; γ-GTP: γ-glutamyl transferase; TBIL: total bilirubin; CRP: C-reactive protein; LDH: lactate dehydrogenase; Glu: glucose; ChE: cholinesterase; CK: creatine kinase; AMY: amylase; BUN: blood urine nitrogen; CRE: creatinine; Na: sodium; K: potassium; Cl: chlorine; Ca: calcium; CEA: carcinoembryonic antigen; CA19-9: carbohydrate antigen 19-9.

Figure 1.

Abdominal contrast-enhanced computed tomography at the portal phase. (a) Enlarged lymph nodes extending from the dorsal aspect of the pancreatic head to the region surrounding the superior mesenteric vein (orange arrowheads). (b) An enlarged para-aortic lymph node (orange arrowheads).

Figure 2.

Esophagogastroduodenoscopy with white light imaging of the papilla of Vater. (a) Ampullary cancer lesion before chemotherapy initiation. (b) Placement of an uncovered self-expandable metallic stent within the tumor during the fourth cycle of maintenance durvalumab monotherapy.

Figure 3.

Histological examination of biopsy specimens from the papilla of Vater tumor before treatment. (a) Hematoxylin and eosin staining showing adenocarcinoma with poorly to well-differentiated features (magnification: ×200). (b) Immunohistochemical staining for MUC1 was negative (magnification: ×200). (c) Immunohistochemical staining for MUC2 was positive (magnification: ×200). (d) Immunohistochemical staining for programmed cell death ligand 1 (SP142) was positive in approximately 10% of tumor cells and 5% of immune cells (magnification: ×200).

Figure 4.

Clinical course of anticancer treatment, computed tomography of para-aortic lymph node metastasis (orange arrowheads), and tumor marker findings (GCD, gemcitabine, cisplatin, and durvalumab; EGD, esophagogastroduodenoscopy). The metastatic lymph nodes, initially measuring approximately 30 mm before treatment initiation, showed marked regression after starting GCD therapy and remained significantly reduced in size 18 months later, measuring approximately 8 mm.

Discussion

Globally, evidence regarding chemotherapy for AC remains limited. The TOPAZ-1 trial, which demonstrated the efficacy of immune checkpoint inhibitors in the treatment of biliary tract cancer for the first time, did not include patients with AC. Conversely, AC is classified as biliary tract cancer in the Japanese guidelines, and chemotherapy regimens for biliary tract cancers have been empirically applied to patients with AC. Given these circumstances, as of 2025, the use of GCD regimens for unresectable AC is covered by health insurance in Japan. To our knowledge, few previous reports have specifically described the clinical course or treatment outcomes of GCD therapy in AC. In two single-center retrospective studies on GCD therapy for biliary tract cancer, only a small number of patients with AC were included.^7,8 Kurihara et al. ⁹ reported that among 52 patients with biliary tract cancer treated with GCD therapy in a multicenter study, three had AC, with a median overall survival of 7.4 months. This case is valuable because it demonstrates the sustained efficacy of GCD therapy for 18 months in a patient with advanced intestinal-type AC. Moreover, PD-L1 expression was evaluated in a pathology specimen obtained before chemotherapy initiation.

The papilla of Vater is anatomically and histologically complex, as it comprises two types of mucosal tissue: the pancreatobiliary ductal mucosa and intestinal mucosa. Consequently, AC is histologically classified into two major subtypes: the pancreatobiliary type and intestinal type. ¹⁰ Immunohistochemical markers such as MUC1, MUC2, CDX2, and CK20 assist in distinguishing these subtypes. The intestinal type is characterized by (1) positive staining for CK20, CDX2, or MUC2 along with negative staining for MUC1 or (2) positive staining for CK20, CDX2, and MUC2 irrespective of the MUC1 result. The pancreatobiliary type is typified by positive staining for MUC1 and negative staining for CDX2 and MUC2. ¹¹ The current National Comprehensive Cancer Network (NCCN) guidelines recommend that chemotherapy for AC is tailored to the histological subtype. For the pancreatobiliary type, regimens aligned with pancreatic or biliary tract cancers are recommended, whereas colorectal cancer-based protocols are generally preferred for the intestinal type. ¹² Conversely, as of 2025, AC is classified as biliary tract cancer in the Japanese guidelines. ⁵ Consequently, GCD therapy was selected as the treatment in our case; however, detailed examination of pathological specimens demonstrated that the tumor was intestinal-type AC. Nevertheless, tumor shrinkage was achieved with the GCD regimen, and this apparently contradictory finding highlights the challenges in selecting treatment options for AC.

Durvalumab is a human monoclonal antibody targeting PD-L1, which functions as an immune checkpoint inhibitor. ¹³ The TOPAZ-1 trial established the clinical efficacy of adding durvalumab to gemcitabine/cisplatin in advanced biliary tract cancers, with a trend toward greater benefit in patients positive for PD-L1. Notably, a benefit was also observed in patients negative for PD-L1, suggesting that the treatment is efficient irrespective of PD-L1 status. ⁶ Conversely, in stage III non-small-cell lung cancer (NSCLC), durvalumab significantly prolonged progression-free survival, predominantly in PD-L1–positive cases following chemoradiotherapy. ¹⁴ These findings suggest that PD-L1 expression can modulate the therapeutic response to durvalumab depending on the tumor type and context. In the present case, PD-L1 expression was confirmed before initiating GCD therapy, and durvalumab may have played a central role in the prolonged treatment response. Alternatively, in NSCLC, immune checkpoint inhibitor administration alters the cancer immune microenvironment and increases tumor sensitivity to chemotherapy. ¹⁵ Durvalumab may have increased the sensitivity of intestinal-type AC to gemcitabine and cisplatin; however, the efficacy of these treatments for intestinal-type AC remains unclear. Several reports have described PD-L1 expression in AC. Dong et al. reported that 34.48% (22/64) of Chinese patients with AC were PD-L1–positive, ¹⁶ whereas Saraggi et al. ¹⁷ detected PD-L1 expression in 26.9% (7/26) of Caucasian patients, with 6/7 having intestinal-type AC. These data suggest that a subset of AC tumors express PD-L1. Interestingly, the TOPAZ-1 trial indicated that PD-L1 expression status had limited predictive value in determining the efficacy of GCD therapy in biliary tract cancers. ⁶ However, as AC was excluded from the TOPAZ-1 trial and is not classified as a biliary tract cancer in the NCCN guidelines, the clinical significance of PD-L1 expression in AC remains unclear. Furthermore, this case involved intestinal-type AC, which, pathologically, shares more similarities with colorectal cancer than with typical biliary tract cancers. Although definitive conclusions cannot be drawn, PD-L1 positivity in this tumor may have contributed to the marked therapeutic response observed with GCD therapy.

In this case, GCD therapy was administered for approximately 9 months, followed by durvalumab monotherapy maintenance for approximately 9 months. The primary therapeutic effect, particularly during the early phase, may have been mainly caused by gemcitabine and cisplatin rather than durvalumab. The long-term impact of durvalumab monotherapy on overall survival may become clearer with continued follow-up. Moreover, microsatellite instability and comprehensive genomic profiling—key biomarkers for predicting immunotherapy benefits—were not assessed because of limited tissue sample availability. This absence represents a significant limitation of this report and underscores the necessity of incorporating molecular profiling in future cases to better guide treatment strategies. According to the NCCN guidelines, chemotherapy regimens used for colorectal cancer are recommended for intestinal-type AC. Therefore, the efficacy of gemcitabine and cisplatin therapy, which is recommended for pancreatobiliary-type AC, for intestinal-type AC remains unclear. Further accumulation of clinical cases and evidence regarding chemotherapy for AC is needed in the future.

The management of cholangitis in AC can also be challenging. In this case, cholangitis attributable to stent obstruction did not occur because of good tumor control. In a multicenter retrospective study focusing on biliary drainage for unresectable AC, Sugiura et al. ¹⁸ reported that SEMSs were associated with a significantly lower rate of recurrent biliary obstruction (RBO) and a significantly longer time to RBO than plastic stents, and no significant differences in outcomes were observed among different types of SEMSs (fully covered, partially covered, and uncovered). For biliary drainage in unresectable malignant distal biliary obstruction, uncovered SEMSs generally carry a lower risk of migration than SEMSs, but the former are more prone to obstruction because of tumor ingrowth. ¹⁹ In the present case, the use of uncovered SEMS resulted in no migration, and effective tumor control achieved through GCD therapy likely contributed to the absence of tumor ingrowth-related obstruction, permitting sustained, long-term effective biliary drainage.

In conclusion, we report a rare and instructive case of intestinal-type AC in which effective disease control was achieved using GCD therapy. Given the current lack of robust evidence regarding chemotherapy in AC, further research and case accumulation are essential. In addition to developing treatment strategies based on histological subtypes, further investigation into PD-L1 expression and the potential applicability of immune checkpoint inhibitors might be necessary. Although the findings of a single case cannot be generalized to all patients with AC, they suggest that durvalumab-containing regimens may offer benefits for selected patients with intestinal-type AC.