MPA involves tumors, such as mucinous adenocarcinoma and signet ring cell carcinoma and produces mucin, which is observed by histological analysis [4]. Both of these two histological types are defined as undifferentiated [3, 25]. Generally, the degree of cancer cell differentiation is associated with cancer aggressiveness. Undifferentiated cancer correlates with aggressive growth, invasion, metastasis and poor prognosis [3, 26].
However, the prognosis of gastric cancer patients with MPA is still controversial and unclear. Several studies have reported that MPA correlates with poor prognosis compared with other histological types [7,8,9,10], while other reports have not yielded significant differences in disease prognosis [11,12,13,14,15]. Moreover, it has also been shown that signet ring cell carcinoma is associated with a favorable prognosis than other types of gastric cancer [19,20,21,22].
Therefore, the present study retrospectively analyzed 1515 MPA gastric cancer patients from the SEER database in order to assess the prognostic value of MPA. In the present study, the proportion of MPA in gastric cancer was approximately 18.1% (14,243/78303), which was considerably higher than that noted in previous studies (2.6–6.6%) [3, 12, 14]. The proportion of CA in gastric cancer was approximately 46.7% (36,602/78303). This evidence suggested that MPA was a rather rare histological type of gastric cancer. The data indicated that patients in the MPA group were younger, more females, more black people, more poor/undifferentiated tumors, later clinical stage, more T3, T4, N2 stages, more surgery and chemotherapy than that of the CA group, which suggested that MPA was an aggressive histological type. No significant differences were noted in the OS and CSS time periods between the MPA and the CA groups, suggesting that gastric cancer patients with MPA had similar prognosis with those with CA. However, for patients with poor/ undifferentiated tumor, clinical stage I, T1 and T2, MPA patients had longer OS time than CA patients. MPA patients had shorter CSS time than CA patients after surgery, but there was no difference in OS time between the two groups. MPA patients without surgery had shorter OS and CSS time than that in CA patients. These results suggested that MPA patients had similar survival time with CA patients on the whole, but MPA was an aggressive histological type in some way. MPA was not a prognostic factor of survival. The results were basically in accordance with those from Zheng et al., who compared the CSS of esophageal cancer patients with MPA or CA and found similar survival time periods between the two groups [27].
With regard to the treatment of MPA, the patients who were treated with surgery or chemotherapy had longer OS and CSS time periods than those without treatment, suggesting that surgery and chemotherapy were effective treatments for early and localized gastric cancer patients with MPA. However, MPA patients treated with radiotherapy exhibited similar OS and CSS time with those without radiotherapy in this study, which indicated that radiotherapy may not be an effective treatment for gastric cancer patients with MPA. However, pertinent information regarding radiotherapy was not available in the SEER database, which could have affected the results. Therefore, further investigations are required to identify the practical and clinical significance of radiotherapy for patients with MPA. Moreover, the therapeutic roles of chemotherapy and radiotherapy for patients with metastatic MPA were not conducted in this study and require further studies.
The present study has its limitations. Firstly, it was a retrospective study and the baseline characteristics of both groups were different, which unavoidably contained selection bias. Secondly, due to the rarity of MPA cases and the exclusion of MPA patients with missing data, the number of patients with MPA used in the present study was low, compared with CA, which may affect the results in a way. Thirdly, the SEER database does not collect several important information, such as the type and extent of surgery, details on chemotherapy and radiotherapy and disease recurrence, which could have affected the results. Therefore, larger studies are required to further determine the clinical and pathological roles of MPA.