UCOGCP has a wide age range of onset and is more commonly observed in middle-aged and elderly people (29). Most studies believe that there is no significant gender difference in the incidence of UCOGCP (5, 30), although some studies showed that it is slightly more common in women (14, 31, 32). In terms of clinical symptoms and signs, UCOGCP is similar to PDAC. Non-specific clinical presentations include abdominal pain, abdominal mass, jaundice, weight loss, etc. Among them, abdominal pain or abdominal discomfort is the most common (29). Some patients do not even have any clinical symptoms. Specific clinical symptoms are related to the location of the tumor and its size. In terms of tumor serological characteristics, tumor markers such as carcinoembryonic antigen (CEA) and CA19–9 are either not significantly elevated or fall within the normal reference range (33). Elevated CA19–9 levels in some UCOGCP patients may be associated with the presence of ductal adenocarcinoma components within the tumor. Some reports indicate an increase in neuron-specific enolase (NSE) levels in certain UCOGCP patients (30).

In clinical practice, the diagnosis of UCOGCP mainly relies on histopathological examination of preoperative biopsies or postoperative specimen and there are no specific imaging findings. Preoperative abdominal ultrasound, contrast-enhanced CT, MRI and other imaging examinations can clearly delineate the size and location of the tumor, which complement each other and contribute to the clear diagnosis of malignant tumors. According to existing reports, UCOGCP mostly presents as cystic-based or mixed cystic-solid mass with delayed enhancement (13, 31, 34), and some cases are purely solid or purely cystic. The masses are frequently accompanied by necrosis and hemorrhage (35). The borders are mostly clear, and the surrounding invasion is not obvious. UCOGCP can occur in any part of the pancreas. Most tumors are located in the head of the pancreas (8, 34).Tumors located in the pancreatic head and neck can result in dilation of the pancreatic duct. It is noteworthy that tumor volume of UCOGCP is usually larger (8), with 80% larger than 5cm and 50% larger than 10cm (33). Analysis of the reasons may be related to the following aspects: (1) Tumors grow quickly; (2) Many tumors are located in the body and tail of the pancreas, with relatively low malignancy and late onset of clinical symptoms; (3) Tumor volume is often accompanied by bleeding which often leads to an increase in tumor size. Endoscopic ultrasound (EUS) is a great tool for the diagnosis of UCOGCP. In EUS examination, PDAC is uniformly hypoechoic, while UCOGCP has uneven echo (36). EUS-guided fine needle aspiration (FNA) and fine needle biopsy (FNB) are able to acquire tumor tissue and further help perform immunohistochemical analysis, which is the only method to diagnose UCOGCP before surgery (8). However, preoperative FNA will increase the incidence of postoperative complications and requires careful selection. Meanwhile EUS-guided FNA can sometimes give the wrong diagnosis. In Muraki’s study, of the 15 UCOGCP patients who underwent preoperative lesion biopsy, only 4 accurate diagnoses were achieved (32). PET-CT can provide useful information about the metabolic status of tumors and the presence of distant metastasis (37). The newer endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) combined with next-generation sequencing technology also facilitates related diagnosis (38, 39).

UCOGCP is an aggressive non-endocrine tumor of pancreas with no glandular differentiation. Histopathological and immunohistochemical classification remains the gold standard for the diagnosis of UCOGCP. Microscopically, its cellular components are mainly composed of neoplastic mononuclear cells and numerous distinctive multinucleated giant cells. These giant cells resemble the giant cells of the bone which is why they are also called osteoclast-like giant cells. The multinucleated OGCs are considered non-neoplastic because they do not share the molecular aberrations of the neoplastic pleiomorphic mononuclear cells. It is often accompanied by massive hemorrhage and necrosis, which may affect the results of pathological diagnosis. OGCs usually have densely eosinophilic cytoplasm and multiple relatively uniform and bland nuclei clustering in the central aspect of the cell (32). Immunohistochemically, OGCs express CD68, does not express cytokeratin AE1/AE3, and has a very low Ki-67 proliferation index (14, 40). Pleomorphic tumor cells are positive for Vimentin, usually express markers of epithelial differentiation (cytokeratin, EMA), may have mutated p53 immuno-profile, and have a high Ki-67 proliferation index (41, 42). Most cases exhibit prominent intraductal/intracystic growth (32). The histogenesis of UCOGCP is still controversial. From the immunohistochemistry results, monocytes showed characteristics of epithelial origin, and OGCs showed characteristics of mesenchymal origin. Most current studies believe that it is of epithelial origin (7, 8). Most UCOGCP are accompanied by KRAS gene mutations, which also supports the origin of ductal epithelial cells (43, 44). There are also reports of lesions originating during epithelial-to-mesenchymal transition (EMT) (45). Through machine learning of gene expression patterns between UCOGCP samples and PDAC samples, it was found that OGCs originate from stem cell-like mesenchymal epithelial cells (SMEC) (46). UCOGCP can be accompanied by other epithelial tumors, PDAC is the most common, and 75% of cases can be accompanied by PDAC at the same time (8, 32). It can also be accompanied by mucinous cystic neoplasm (MCN), intraductal papillary mucinous neoplasm (IPMN) or malignant stromal tumors (15, 31, 47, 48). Table 2 shows literature reports on UCOGCP with mixed components. If no other distinct epithelial tumor components can be found, it’s called pure UCOGCP. Through analysis of the immunohistochemical expression of three well-known EMT markers, EMT activation was found more frequent in anaplastic carcinomas than in UCOGCP (58).

Modern gene sequencing technologies such as next-generation sequencing (NGS) have been widely used. In addition to long-term follow-up, investigation of the genomic alterations in confirmed and surviving patients can also help guide the selection of treatment options and prognostic analysis. To better understand UCOGCP, many genetic features have been identified. The most frequently mentioned gene mutation of UCOGCP in literature reports is the KRAS gene (59, 61). Most KRAS mutations were found in codon 12, including the G12V, G12D and G12R mutations. KRAS is one of the most common oncogenes in human cancer. KRAS activation is an early event in the pathogenesis of pancreatic cancer. KRAS mutations are present in approximately ~85% of PDAC patients (62). Whole-exome sequencing (WES) of eight cases showed that UCOGCP has surprisingly similar molecular characteristics compared with conventional PDAC (54). In addition to the common KRAS oncogenic mutations, other typical mutated genes including TP53, CDKN2A and SMAD4 were found in UCOGCP, which further confirm that UCOGCP is a unique variant of PDAC. Mutations in SERPINA3, GLI3, MAGEB4, MEGF8 and TTN have also been detected. Some of them may act as pathogenic or likely pathogenic mutations. Zhao et al. detected 6 potentially clinically significant genes through NGS: AR, FBXW7, CCNE1, BTK, KRAS, TP53 (48). Somatic BRCA2 alterations were also detected by WES in one case (6). Recently, Hrudka et al. performed molecular genetic analysis of 13 UCOGCP cases with varying survival time (60). This is the largest NGS examined UCOGCP cohort to date. The results showed that the mutant spectrum of UCOGCP is very similar to that of PDAC. Unfortunately, UCOGCP-specific genomic signatures remain mainly unknown and no somatic genetic aberrations were found that were clearly related to patient prognosis or PD-L1 status. Gene characteristics of UCOGCP in literatures are summarized in Table 3 . MicroRNAs (miRNAs) play a variety of important regulatory roles in both healthy and cancer cells. Popov et al. evaluated the expression of seven previously clinically significant miRNAs in UCOGCP and poorly differentiated (grade 3, G3) PDACs (65). Except for miR-155, there were no significant differences in the expression of other miRNAs in the two tumors.

Regarding UCOGCP, because its incidence is very low, most literatures are case reports, there are few randomized controlled studies of large cohorts, and there are no relevant guidelines for reference. Consequently, the clinical course of this disease remains poorly understood. Overall, radical surgical resection with negative margins (R0 resection) is the optimal treatment when surgical indications are met. The selection of surgical method and the scope of lymph node dissection can refer to PDAC, which mainly depends on the location of the tumor, regional lymph node involvement and intraoperative exploration. If nearby organs such as the stomach, jejunum, colon, and kidneys are invaded, they may need to be resected together. Because this type of tumor has large lesions and rich blood supply, careful preoperative imaging evaluation can help improve the complete resection rate and safety of surgery.

Pancreatic cancer is the most aggressive solid tumor in humans with a very poor prognosis. The treatment of pancreatic cancer is based on tumor histology. Because UCOGCP is so rare, the optimal treatment approach for this kind of tumor has not yet been determined. Several randomized clinical trials have confirmed that patients with PDAC at all stages need to receive chemotherapy. Since UCOGCP is a variant of pancreatic ductal adenocarcinoma, the common chemotherapy regimen for PDAC such as FOLFIRINOX(combination of oxaliplatin, irinotecan, fluorouracil and leucovorin) and gemcitabine can be often seen in case reports of UCOGCP. On the other hand, it has been reported in the literature that the comprehensive treatment strategy for UCOGCP can refer to undifferentiated pancreatic cancer. And a retrospective multicenter cohort study suggests that for patients with UCP, nab-paclitaxel-based combination chemotherapy may be more advantageous and significantly improved OS compared with non-paclitaxel-containing regimens (6.94 months vs. 3.75 months, respectively; P = 0.041) (66). For unresectable locally advanced or borderline resectable UCOGCP, there are reports in the literature that FOLFIRINOX regimen and modified FOLFIRINOX can be used for downstaging before surgery (18, 24). Neoadjuvant chemotherapy (NACT) can help reduce tumor size and alleviate local progression, thereby creating opportunities for R0 resection of the tumor without the need for resection of major blood vessels or adjacent organs. And it can also select patients with good tumor biological characteristics, who will benefit from curative resection surgery.

For PDAC, immunotherapy regimens have always had poor efficacy, while there are relatively few reports on the application of immunotherapy in UCOGCP. UCOGCP seems to exhibit high expression of universal programmed death ligand-1 (PD-L1), which may be used as an effective prognostic marker (41, 42, 67). PD-L1 expression was more frequent in UC compared with conventional PDAC (63% vs 15%, P < 0.01) (67). For UCOGCP, PD-L1 was expressed on tumor cells in 17 of 27 cases (63%) and was more often in patients associated with PDAC (42). Moreover, multivariate analysis confirmed that PD-L1 expression was associated with poor prognosis (HR = 3.397; 95% CI, 1.023–18.375; P = 0.034). The reason behind this may be that PD-L1 suppresses anti-tumor immunity and allows tumor cells to evade the cytotoxic activity of host T lymphocytes.

Another study including 13 UCOGC patients demonstrated similar conclusions (41). UCOGC expressed PD-L1 significantly more frequently and had greater numbers of CD3+ and CD8+ tumor-infiltrating lymphocytes (TILs) compared with PDAC. Among the 3 PD-L1 negative UCOGCP cases, none of them reached the death end point. The median survival time of PD-L1-positive patients was 5.7 months. However, due to the small cohort size, the comparison was not statistically significant. Immunotherapy may exert anti-tumor effects on UCOGCP with distant metastasis. Obayashi et al. (68) reported a UCOGCP patient with pancreatic body and tail tumors and lung metastases who achieved clinical complete remission of lung metastases after monotherapy with pembrolizumab monoclonal antibody. The patient achieved disease-free survival after subsequent resection of the pancreatic body and tail tumors. For a patient with unresectable UCOGCP with lung metastases, whose tumor mutation burden (TMB) was high as measured by NGS, he started to receive third-line pembrolizumab monotherapy after palliative radiotherapy. After 32 months of treatment, the primary tumor and metastases achieved sustained remission (16). So far, the effectiveness of immunotherapy in patients with UCOGCP still needs to be verified in large cohorts and more clinical trials.

Reports on UCOGCP are scarce and the terminology and classification of UCOGCP are not uniformly standardized. As a result, the prognostic data is fairly limited and vary greatly. Large survival differences may be related to tumor tissue heterogeneity and different tumor stages at diagnosis. Overall, UCOGCP is a highly aggressive tumor. UCOGCP lacks typical clinical symptoms and imaging features, and has poor tissue differentiation and strong invasiveness. UCOGCP is often diagnosed in the advanced stages with large tumor size, and many patients cannot get complete surgical resection. There is also a lack of effective therapeutic drugs. Even if it can be resected, early recurrences are also common. So the overall prognosis is poor. The opinion given in the WHO 2019 Digestive System Tumors Volume is that patients may survive for 1 year or more (9). The survival time of UCOGCP reported in most existing literature also shows similar or better results (30, 41, 69). Based on the data analysis of UCP patients in the National Cancer Database (NCDB), the median survival time of UCOGCP patients was 24.8 months and the estimate 3-year and 5-year survival rates were 42.54% and 39.01%, respectively (70). The difference in survival is also related to the stage of the tumor. By integrating previous literature reports, the median overall survival time of UCOGCP patients who underwent surgical resection was found to be 48 months (29). For unresectable UCOGCP patients with locally advanced tumors or distant metastasis, the prognosis is significantly worse and median survival time is less than 6 months (71, 72). Table 4 lists more detailed prognostic data of UCOGCP based on reported case series.

Pure UCP is highly malignant and has a poor prognosis. It is worth noting that UCOGCP is slightly less malignant than conventional PDAC and undifferentiated carcinoma without OGCs. In Strobel’s study, median survival for patients with anaplastic pancreatic cancer was 5.7 months, compared with 15.7 months for control patients with pancreatic ductal adenocarcinoma (75). The survival times of the three UCOGCP patients were even significantly longer, 33, 49, and 161 months respectively. The tumor resection rate of UCOGCP was also unexpectedly found to be higher than other types of UCP (29). In a recent large database review of UCP, patients with OGCs (UCOGCP) had longer median OS (aHR 0.52: 95% CI 0.41–0.67) compared with patients without OGCs (70). A proportion of UCOGCP patients in previous publications are still alive many years after diagnosis. A patient with locally advanced UCOGCP survived for even 10 years after undergoing surgery and adjuvant therapy (33). In one literature report, the patient’s survival time was even 15 years (76). Despite its rapid growth and large tumor size, the biological behavior of UCOGCP is different from that of common PDAC (77). Although UCOGCP grows rapidly, recent reports suggest that perineural invasion and lymph node metastasis are rare, and its prognosis is relatively better than that of PDAC and pleomorphic giant cell tumor (32, 78). According to Muraki’s study, the five-year survival rate of resected UCOGCP is 59.1%, while the five-year survival rate of resected PDAC is 15.7% (32). However, some research results showed that the prognosis of UCOGCP patients may be worse than that of PDAC patients, and there is a lack of effective large sample data to further clarify this issue (30).

However, some literature demonstrated different results. One retrospective study evaluated 55 patients with unresectable UCP. Final survival analysis showed that there is no significant difference in OS between the UC with OGCs group and the UC without OGCs group (72). Apart from tumor stages at diagnosis, the possible reasons behind may be related to the pathological components and the different roles of OGCs. More detailed pathological differences may have an impact on the prognosis of UCOGCP patients. As mentioned before, UCOGCP can be categorized into 2 distinct pathological subtypes: pure UCOGCP, and UCOGCP associated with other epithelial tumors. Patients with UCOGCP alone seem to survive longer than patients with UCOGCP accompanied by PDAC (30, 32, 54, 79). A meta-analysis using previous reported data showed that major prognostic factors of UCOGCP include age, sex, tumor size, lymph node metastasis, and concomitant PDAC component (79). However, concomitant MCN component did not affect prognosis.