Continuous Regression of Metastatic Pancreatic Adenocarcinoma after Suspending Chemotherapy: A Case Report

Abstract

Introduction

Regression of malignancy in the absence of cancer-directed therapy is an uncommon and poorly understood phenomenon. Its occurrence is particularly rare for pancreatic ductal adenocarcinoma (PDAC). The interaction between tumor and microenvironment may induce an immune response in which both innate and acquired immunity have been found to be implicated. An immunogenic tumor may promote antigen presentation and effector T-cell activity, leading to cancer cell death and tumor inhibition.

Case Presentation

We present the case of a 57-year-old man diagnosed 3 years ago with a borderline-resectable PDAC confirmed by biopsy. The patient received first-line 5-fluorouracil-based chemotherapy with progression of disease including new hepatic metastases. He then received twelve cycles of gemcitabine and nab-paclitaxel with successful reduction in the number and size of liver metastases. Upon patient request, a treatment holiday was initiated, during which his abdominal imaging showed continued tumor regression. After 11 months without any systemic therapy, there are no remaining metastases in the liver, and the primary pancreatic mass continues to recede. He currently remains on surveillance.

Conclusion

Our patient’s rare clinical course raises questions including the optimal next steps in treatment, such as continued observation or local treatment such as surgical resection. The decision to continue observation results from our belief that minimal disruption of the tumor microenvironment may allow for continued control and cancer regression. More laboratory and clinical studies are imperative to understanding the physiological basis of sustained tumor regression after chemotherapy discontinuation and may impact real-world clinical decision-making. Additionally, the definition of spontaneous tumor regression may need revision to distinguish between prolonged therapeutic response and regression due to immunogenic features in the absence of prior therapy.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) has an aggressive disease course with very poor prognosis. It accounts for 3.3% of all new cancer cases in the USA in 2025, with about 67,400 individuals diagnosed annually. Approximately 51,980 individuals have died of PDAC this year, comprising 8.4% of all cancer-related deaths. Based on analysis of data between 2018 and 2021 from the Surveillance, Epidemiology, and End Results (SEER) Program, the lifetime risk of developing this disease is about 1.6% in the general population; between 2015 and 2021, the 5-year relative survival rate was only 13.3% [1].

Spontaneous regression of malignancy was defined in 1956 as the partial or complete disappearance of a neoplasm without treatment, or in the presence of therapy that is deemed inadequate to exert control over tumor growth [2]. The occurrence is poorly understood and rarely documented. Spontaneous regression occurs in approximately 1.0 to 1.6 per 100,000 cancer patients [2]. While certain malignancies, such as melanoma or infant neuroblastoma, have been reported to spontaneously regress in a minority of patients, such a phenomenon in PDAC is among the rarest [3].

Dr. William Coley, referred to as the Father of Immunotherapy, noted an association between infection and tumor regression, leading him to produce and inject “Coley’s toxin” into inoperable sarcomas as a form of cancer therapy [4]. These observations led to the theory that a patient’s immune response to perceived foreign antigens on tumor cells may cause tumor regression, while a secondary source of inflammation may prime the immune system to antigen recognition.

Spontaneous regression of PDAC has been rarely described. They often identify a concurrent source of inflammation. In the first case described in 1934, a patient underwent exploratory laparotomy with biopsy confirming PDAC. She had a complicated 2-month surgical recovery and never received cancer-directed therapy. She passed away 7 years later of pulmonary embolism, but autopsy did not find any trace of cancer [5]. In another case, prior to pancreaticoduodenectomy, the patient suffered a non-ST-elevated myocardial infarction requiring prolonged hospitalization. Subsequent imaging verified sustained tumor regression and carbohydrate antigen 19-9 (CA 19-9) normalized [6]. In a third report, a patient with PDAC developed a perforated duodenal ulcer, peritonitis, and pneumonia after receiving insufficient treatment with 2 weeks of radiation therapy and 7 weeks of gemcitabine without treatment response. Scans and CA 19-9 levels during and after admission confirmed tumor regression [7].

One prior case reported a patient who had previously received adequate treatment for metastatic PDAC with 6 months of folinic acid, fluorouracil, irinotecan, and oxaliplatin (FOLFIRINOX). Due to significant toxicity, the patient elected to stop further treatment. Subsequent scans confirmed continued response in tumor size and CA 19-9, and the patient ultimately achieved complete regression off of therapy which was sustained for 33 months [8]. The clinical findings were reported as spontaneous tumor regression.

In this report, we describe a case of PDAC with prolonged tumor regression after the completion of systemic therapy. In our patient’s clinical course, there was no secondary source of inflammation. Based on our search, this is the second documented case of PDAC with a prolonged and continuous therapeutic response almost 1 year after treatment discontinuation. Our patient’s molecular sequencing did not suggest tumor immunogenicity.

Case Description

Our patient was 54 years old when he was first diagnosed with PDAC. He sought medical attention due to jaundice, abdominal pain, and weight loss. His medical history includes gastritis with Helicobacter pylori infection which was treated. Our patient also has a ten pack-year cigarette smoking history and quit tobacco use at the age of 40. He has no history of significant alcohol use and no illicit drug use. He immigrated from China with his family about 25 years ago.

His blood work revealed elevation in his liver enzymes and hyperbilirubinemia: his alanine aminotransferase was 532 U/L, aspartate aminotransferase was 370 U/L, alkaline phosphatase was 497 U/L, total bilirubin was 10.8 mg/dL, and direct bilirubin was 7.6 mg/dL. He subsequently underwent a contrast-enhanced computed tomography (CT) of the abdomen and pelvis which showed a 3.7-cm mass at the head of the pancreas. The mass abutted, but did not encase, the superior mesenteric artery (SMA) and superior mesenteric vein (SMV); the remaining abdominopelvic vasculature was free of disease. There were no visible abdominopelvic metastases (Fig. 1).

Fig. 1.

Contrast-enhanced CT scan at time of diagnosis. A hypoattenuating pancreatic head mass measuring 2.8 × 2.0 × 3.7 cm (TR × AP × CC) (arrowheads, a, b) highly suspicious for pancreatic adenocarcinoma demonstrates significant biliary obstruction (arrows, c). No abdominopelvic metastasis was radiologically present at time of diagnosis.

An endoscopic retrograde cholangiopancreatography exam was performed with gastric and pancreatic biopsies obtained and one plastic stent placed in the common bile duct. Endoscopic ultrasound-guided fine-needle aspiration confirmed a diagnosis of PDAC. Genetic testing confirmed the absence of a hereditary cancer predisposition.

After stent placement, there was improvement in our patient’s transaminase levels and normalization of his bilirubin. Due to his excellent performance status and borderline-resectable tumor based on surgical evaluation, he began treatment with FOLFIRINOX. He continued treatment for a total of twelve cycles for a duration of 6 months. This regimen was discontinued due to evidence of disease progression on his CT scans, which demonstrated an interval increase in the size of the pancreatic head mass with continued less than 90-degree abutment of the SMA, but new near-complete occlusion of the portal confluence and severe narrowing of the proximal SMV. There was also evidence of multiple new liver metastases, mostly in the right hepatic lobe, with the largest lesion measuring 1.1 cm. At least twelve to fourteen lesions were visualized (Fig. 2).

Fig. 2.

Contrast-enhanced CT scan 6 months after diagnosis. The pancreatic head mass slightly increased in size, measuring 3.5 × 2.9 × 3.9 cm (TR × AP × CC) (arrowheads, a, b), but with new severe narrowing of the SMV (open arrowhead, a). New hypoattenuating liver lesions were highly suspicious for hepatic metastasis; the largest measured 1.1 cm in the peripheral right lobe (arrow, c).

Due to the new imaging findings, within 1 month our patient was sent for liver mass biopsy by interventional radiology. During the procedure, the hepatic lesions could not be visualized on intraprocedural CT scan or ultrasound, and biopsy could not be obtained. These findings, however, coupled with doubling of the serum CA 19-9 during the same period from 66 to 133 U/mL led the treatment team to consider these as true metastases. Furthermore, the hepatic metastases increased in size on subsequent CT scan, indicating definite radiologic metastasis (Fig. 3). The patient continued to have excellent performance status. He was therefore switched to second-line treatment with gemcitabine and nab-paclitaxel. A liquid biopsy was attempted just prior to starting the second regimen; however, tumor DNA quantity was insufficient in the blood sample for next-generation sequencing (NGS).

Fig. 3.

Contrast-enhanced CT scan 12 months after diagnosis. The pancreatic head mass remains unchanged in size (arrowheads, a, b). Significant interval growth of the right lobe hepatic metastasis, now measuring 1.8 cm (arrow, c), is consistent with radiologic disease progression.

Our patient’s disease response continued to be monitored while he received the second chemotherapy regimen. Due to significant reduction of disease burden, he received a total of twelve cycles of this regimen. Due to the prolonged duration of treatment with evidence of reduction of the size and number of liver metastases, the patient elected to begin a treatment holiday with frequent surveillance CT scans. After discontinuing therapy, his CA 19-9 subsequently decreased to within the normal range of 10–14 U/mL.

Three months after discontinuing chemotherapy, his CT scan showed an unchanged size of the 2.4-cm pancreatic head mass with less than 180-degree abutment of the proximal SMV and less than 180-degree abutment of the SMA. Magnetic resonance imaging (MRI) of the abdomen showed continued reduction in the size of all hepatic metastases, with only three residual measurable lesions still visible out of the initial twelve to fourteen lesions, all confined to the right hepatic lobe.

Six months after discontinuing chemotherapy, our patient’s CT scan showed further shrinkage of the pancreatic head mass to 2.0 cm, and only one remaining visible liver metastasis 3 mm in size. MRI findings were consistent with these results (Fig. 4).

Fig. 4.

Contrast-enhanced CT scan almost 3 years after diagnosis. The pancreatic head mass has significantly decreased, now measuring 2.0 × 1.3 × 1.4 cm, and is difficult to distinguish from the adjacent duodenum (arrowheads, a, b). The hepatic metastasis has nearly completely resolved, now present as subtle capsular retraction (arrow, c).

After 10 months off of all cancer-directed therapy, the solitary remaining tumor at the pancreatic head shows no sign of growth and his CA 19-9 remains stable at 14 U/mL. Our patient remains on surveillance with blood work, physical examination, and CT and MRI imaging every 3 months. The option of surgical resection of the remaining sites of cancer in the pancreas and the right hepatic lobe was discussed in two separate multidisciplinary conferences. The consensus in both was that, given the suspicion that our patient’s immune system is effectively controlling tumor growth, surgery should not be pursued to avoid disruption of this mechanism.

Our patient’s previous germline genetic testing was negative for a hereditary cancer predisposition. The results of NGS from his initial biopsy specimen revealed the somatic mutations listed in Table 1.

Table 1.

NGS results from pancreatic tumor fine-needle aspiration specimen at the time of diagnosis

Gene	Somatic mutation
BRCA1	Rearrangement intron 2
ROS1-DCLBD1	Noncanonical fusion
CDKN2A	Loss
KRAS	G12D point mutation
MTAP	Loss
CDKN2B	Loss
KEL	Splice site 1413+1G>C
SMAD4	Loss exons 5–9
TP53	R248Q

Discussion

Spontaneous regression of PDAC is a rare clinical event, with only isolated cases reported in literature. This phenomenon contrasts with the typical clinical course of PDAC, characterized by rapid progression, early metastasis, and chemotherapy resistance. Here, we report the case of a patient with metastatic PDAC who demonstrated sustained regression of both the primary tumor and liver metastases despite having discontinued systemic therapy. This occurred in the context of an unfavorable genomic profile, without exposure to an immune checkpoint inhibitor, and in the absence of an overt immune-triggering event. These aspects of our patient’s disease create a compelling case for exploring noncanonical drivers of tumor regression, particularly within the tumor microenvironment (TME).

At the molecular level, the patient’s tumor exhibited classical alterations: KRAS G12D, TP53 R248Q, SMAD4 loss (exons 5–9), and biallelic deletion of CDKN2A, CDKN2B, and MTAP. These mutations are associated with oncogenic signaling, genomic instability, defective cell cycle regulation, and evasion of TGF-β-mediated growth suppression [9, 10]. Additionally, a BRCA1 intron 2 rearrangement was identified, raising initial suspicion of homologous recombination deficiency, a feature that can enhance neoantigen generation and immune surveillance [11]. However, functional homologous recombination deficiency testing was negative, diminishing the likelihood that DNA repair impairment played a dominant role in this case.

Importantly, this patient’s tumor lacked genomic hallmarks typically associated with immunogenic cancers. NGS revealed that the tumor was microsatellite stable, POLE wild-type, and exhibited low tumor mutational burden (TMB 0 Muts/Mb), features that correlate with poor immunogenicity. The absence of these biomarkers makes his clinical course particularly remarkable, since tumors with such characteristics are generally resistant to immune activation and checkpoint blockade. This suggests that tumor regression resulted from noncanonical mechanisms of immune activation rather than from intrinsic genomic immunogenicity.

Most previous reports of spontaneous PDAC regression occurred in the setting of secondary inflammation. In the one reported case of PDAC regression in which the patient had received 6 months of FOLFIRINOX, the patient’s molecular testing had revealed KRAS G12D mutation, CDKN2A deletion exons 1–3, CDKN2B exon 1 truncation, TP53 R273C mutation, and the tumor was microsatellite stable with a low TMB [8]. This patient’s findings were reported as spontaneous tumor regression. Her tumor molecular profile was thus similar to that of our patient with poor immunogenicity.

Although the tumor molecular profile of our patient suggested a poor prognosis and immune-evasive phenotype, the clinical course diverged dramatically, thus highlighting the limitations of genomics alone in predicting therapeutic response. PDAC is embedded within a dense, desmoplastic TME composed of cancer-associated fibroblasts, pancreatic stellate cells, immune cells, and extracellular matrix proteins. The TME promotes tumor growth, therapeutic resistance, and immune exclusion. In PDAC, stromal components can constitute up to 90% of the tumor mass. Their density hinders immune cell infiltration, while their biochemical output, TGF-β, IL-6, and CXCL12 among others, creates an immunosuppressive environment [12, 13].

The immune composition of the PDAC microenvironment can be broadly categorized into immune-cold, immune-hot, and immune-exhausted states. Cold tumors are dominated by Tregs and M2 macrophages with sparse CD8⁺ T cells, while hot tumors exhibit strong effector T-cell infiltration, higher neoantigen burden, and active interferon signaling. In contrast, exhausted tumors contain immune cells rendered inactive by chronic antigen exposure and checkpoint ligand expression (PD-L1, TIM-3), leading to resistance to immune- and therapy-driven control [14]. However, the immune states of pancreatic tumors can be dynamic during the disease course. Immune-cold tumors may shift to an immune-active phenotype in response to acute inflammation, vascular remodeling, or chemotherapy-induced immunogenic cell death. Conversely, immune-rich tumors may become exhausted under chronic pressure. This dynamic evolution may explain the rare but significant regressions observed in some PDAC patients following infection, ischemic injury, or other systemic stressors [14].

In our case, although no overt trigger such as infection or febrile illness was identified, the patient’s prior exposure to cytotoxic chemotherapy (FOLFIRINOX followed by gemcitabine and nab-paclitaxel) may have primed the immune system or altered stromal composition. Studies have found that gemcitabine, in addition to be cytotoxic, may also be immunomodulatory. It has been reported to enhance antitumor immunity through mechanisms like depleting Tregs and MDSCs, improving the effector/Treg ratio, augmenting NK and CD8⁺ T-cell activity, and upregulating MHC-I and calreticulin expression on tumor cells, promoting antigen presentation and phagocytic clearance. However, gemcitabine can also promote M2 macrophage infiltration, monocyte expansion, and systemic myelosuppression, producing an immunosuppressive microenvironment [15] Therefore, its immunological effects seem to be context-dependent.

Conclusion

In summary, this is a case of metastatic PDAC with sustained regression of both the primary tumor and liver metastases despite discontinuation of systemic therapy. This is a unique case in which a secondary source of inflammation was absent, and molecular sequencing data were available. Given our suspicion of effective immune control of tumor growth, we favor active surveillance rather than surgical intervention. Additionally, we conclude that the definition of spontaneous tumor regression may need revision to distinguish between prolonged therapeutic response and regression due to immunogenic features in the absence of prior therapy.

A clearer understanding of the interaction between pancreatic tumors and the TME will aid in understanding the physiological basis of tumor regression after treatment discontinuation and may impact real-world clinical decision-making. The CARE Checklist has been completed by the authors for this case report, attached as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000549360).

Statement of Ethics

This study was performed in accordance with the Declaration of Helsinki. Ethical approval for retrospective review of patient data was not required in accordance with local and national guidelines. All adult participants provided written informed consent to participate in this study. Written informed consent was obtained from the individual for publication of the details of their medical case and any accompanying images. Written informed consent was obtained from the patient for publication of this case report and all accompanying images.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

The authors declares no funding was obtained or used for this study.

Author Contributions

Leyla Bayat, MD: primary writer of the report; substantial contributor to conceptualization of the report with delineation of our goals and aims; primary contributor to data acquisition, analysis, and interpretation relevant to the work; final approval of the version to be published; and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Arturo Orlacchio, PhD: contributor to the review and editing of the report with commentary and revision; substantial contributor to the analysis and interpretation of data relevant to the work; final approval of the version to be published; and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Daniel Parrott, MD: contributor to the review and editing of the report with commentary and revision; contributor to the acquisition and interpretation of visual data relevant to the work; substantial contributor to the analysis and interpretation of data relevant to the work; final approval of the version to be published; and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Jennifer Wu, MD: supervisory role with oversight of the report and research; substantial contributor to conceptualization of the report with delineation of our goals and aims; primary critical reviewer of the work for important intellectual content; final approval of the version to be published; and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding Statement

The authors declares no funding was obtained or used for this study.

Data Availability Statement

The data that support the findings of this report are not publicly available to maintain the privacy of the patient. They are available from Leyla Bayat, MD, or from Jennifer Wu, MD, upon reasonable request.