Gallbladder cancer (GBC) is the most prevalent malignant tumor of the biliary system and the sixth most prevalent malignant tumor of the digestive tract (1–3). Adenocarcinomas originating from the secretory cells of the gallbladder are the most common GBC subtype and account for more than 90% of the cases (4–6). Other rare adenocarcinomas include gallbladder papillary adenocarcinoma (GBPA), gallbladder mucinous adenocarcinoma (GBMA), and gallbladder tubular adenocarcinoma (GBTA) (4, 7). GBPA originates from the papillary cells that facilitate the movement of bile in the gallbladder. As the tumor grows, it blocks the flow of bile in the gallbladder. If the tumor blocks the neck of the gallbladder, it can lead to an enlargement of the gallbladder, thinning of the gallbladder wall, and ultimately the formation of an abscess or effusion. In addition, GBPA can also form ulcers on the tumor surface, which are prone to infection. GBPA is often found incidentally in pathological examinations of postoperative specimens. To our knowledge, few studies have been published evaluating the clinicopathological characteristics of GBPA, and most studies are limited to individual case reports or small retrospective series (8–13).

Previous studies reported that the prognosis of GBPA is generally better than the other types of GBC. Veeravich et al. (14) compared the survival of GBPA with other types of GBC using population data obtained from the Surveillance, Epidemiology, and End Results (SEER) database. The conclusion demonstrated that the survival rate of GBPA was much higher than that of GBA and other types of GBC. Although some studies (10, 15, 16) have also confirmed that the prognosis of GBPA is better than other types of GBC, these studies are more than 10 years old and did not perform a propensity score matching (PSM) analysis to evaluate the impact of other various variables on survival. Therefore this study was designed to retrospectively evaluate the clinical factors and prognosis of patients diagnosed with GBPA, applying data obtained from the SEER database. The objectives of the studies were to: (1) represent and contrast the characteristics of GBPA and other types of GBC (GBA, GBMA); (2) perform a propensity score matching (PSM) analysis to compare the survival rate of GBPA and other types of GBC to account for variations in cofounding factors; (3) develop and evaluate the price of prognostic nomograms for patients with GBPA.

In 2020, there will be an estimated 115949 new GBC cases worldwide, ranking 25th out of 36 cancers in 185 countries (17). GBPA is a rare subtype of GBC. Due to its low incidence rate (14), the OS and CSS are still unclear as large sample cohort studies are lacking (9, 10). As a result, further research is required to evaluate the clinical significance of this GBC subtype. To our knowledge, there is no relevant literature comparing survival rates (OS and CSS) of GBPA with other GBC subtypes after PSM. Therefore, the first population-based large-sample research aiming to build nomograms to predict survival rates (OS and CSS) of GBPA based on clinical, demographic, and outcome data obtained from the SEER database.

Previous studies have shown that GBC is the only cancer in the digestive system that is more common in women than men (18, 19). Consistent with previous studies, we also found that all GBC subtypes are more common in women. However, the ratio of female of GBPA was significantly higher than that of GBMA. Similar to previous studies, more than 70% of the patients with the GBC subtypes were aged 60 years and above (20). Compared with GBA, the GBPA subtype was less common in white patients, while no significant racial differences were noted between the GBPA and GBMA subtypes.

Studies have also shown that GPBA has unique pathological characteristics. Consistent with our findings, Wan et al. (10) also showed that the differentiation degree and TNM stage between the GBPA and GBA subtypes differed. Zou et al. (21) also showed a difference in the TNM staging between GBA patients and GBMA patients, but no comparison was made between GBPA and GBMA patients.

One of the previous research (10) have also confirmed that the prognosis of GBPA is better than other types of GBC and the 1-year, 3-year, 5-year, and 10-year OS of patients with GBPA were 100%, 76.9%, 76.9%, and 38.5%, respectively. However, the study based on Chinese data has been almost 10 years old and the sample of patients with GBPA is only 16. There were 281 GBPA patients in our study and showed that the 1-year, 3-year, 5-year, and 10-year OS of patients with GBPA were 81.3%, 58.8%, 51.7%, and 34.2% before PSM, while the 1-year, 3-year, 5-year, and 10-year OS of patients with GBPA were 80.6%, 56.3%, 47.2%, and 33.3% after PSM. Our research results have a sufficiently large sample and comparative analysis before and after PSM, which to some extent avoids bias in the study, so the results obtained may be more reliable.

In this study, PSM analysis (ratio 1:1) was used to represent survival status differences between GBPA and other forms of GBC (GBA and GBMA). In line with previous reports (14, 16, 18, 22, 23), the data before PSM indicated that patients with the GBPA subtype had a higher OS and CSS than patients with the GBA and GBMA subtypes. However, after applying PSM, GBPA still had a higher OS than GBA, but no statistical difference was noted in the CSS. Meanwhile, the OS and CSS of GBPA and GBMA remained significantly different.

The prognosis of patients with GBPA is relatively good, which may be related to the inherent characteristics of this tumor. Compared with other types of GBC, the ability of GBPA to infiltrate the gallbladder wall is reduced, which to some extent delays the progression of the disease. In addition, Zou et al. (10) have shown that obstructive symptoms related to GBPA lesions may occur earlier, which helps early detection and treatment of GBPA patients and to some extent, improves the prognosis of GBPA patients.

Cox proportional hazard analysis was used to determine prognostic factors in patients with GBPA. Age, marital, T stage, and M stage were recognized as independent prognosis factors of OS in GBPA patients, while T-stage, M-stage, and surgery were recognized as independent prognosis factors of CSS in GBPA patients. These findings are consistent with the study of Wan et al. (10), which showed that jaundice, T-stage, and nodal involvement were independent predictors for OS in patients with GBPA. More than 90% of GBPA patients in this study had no distant metastasis. As a result, most of the GBPA patients (97%) in our study were treated with surgery. Previous reports (24–26) have shown that surgery is an effective treatment for GBC, but there are some controversies about the prognosis. Our findings indicate that surgical treatment in GBPA patients can improve CSS but not the OS. Conversely, chemotherapy and radiotherapy were not identified as independent prognosis factors for OS and CSS in patients with GBPA.

The independent prognosis factors were used to establish predictive nomograms (OS and CSS), specifically for patients with the GBPA subtype. The ROC curve and calibration chart showed that our proposed nomogram has good prediction ability. Furthermore, the DCA showed that the application of this nomogram in clinical practice provided a positive net clinical benefit compared to the standard TNM staging system.

Our study has some limitations that have to be acknowledged. Owing to a population-based retrospective analysis based on data available on the SEER database, important variables, such as preoperative laboratory results and comorbidities, could not be analyzed and incorporated into our prediction model. Finally, all the data in our study were obtained from a single open-source database. Therefore before applying this model clinically, external validation is required.

In the study, we carried out a population-based study using the SEER database to demonstrate the clinicopathological features and survival of patients with GBPA. The results demonstrated that GBPA is an unusual GBC with unique characteristics and prognosis. In addition, we have also establishmented nomograms to predict survival rates (OS and CSS) of GBPA patients. The established nomograms provided a better survival prediction for GBPA patients than the traditional TNM model. The nomograms could be used to guide clinicians in conducting personalized diagnosis and treatment of GBPA patients. However, further validation of external data is needed to assess the generalizability of the findings.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

ZW, HW and LW designed and performed the study. HW, XZ provided administrative support. ZW, LW, YH and XZ collected and assembled the data. YB, ZW contributed to data processing. ZW, LW and YB prepared all the tables and figures, interpretation of results, and drafting. All authors reviewed and approved the manuscript. All authors contributed to the article and approved the submitted version.