Abstract
The number of elderly patients in gastric cancer surgery is rapidly rising. Almost all data on gastric cancer in people over the age of 80 come from a single institution, and there is no systematic review of a large number of patients. Therefore, we conducted a comprehensive analysis of the prognosis of patients with gastric cancer surgery who were aged 80 years or older. From January 2010 to November 2021, reports on gastric cancer in the elderly aged 80 and over were gathered. We searched PubMed for “Gastric cancer and elderly and 80 years old” as a keyword, and 253 reports were extracted. The Ichushi-Web database was also searched using the phrase “stomach cancer and 80 years old,” and 366 records were found. The random-effect model was used to determine the average 5-year survival rate, and the heterogeneity was evaluated. The proportion of male patients, patients who had surgery after 2010, patients with stage I, total gastrectomy, lymph node dissection, and the presence of complications were used as the explanatory variables in meta-regressions to investigate the cause of prognosis variability. More than 50 surgical cases were reported, 8 from PubMed and 2 from the Ichushi-Web database, with information on surgical procedures, prognosis, and complications, in a total of 1182 patients. Of the ten reports, eight were from Japan and two were from South Korea and Taiwan. The number of patients ranged from 55 to 217, with an average 5-year survival rate of 57%. In terms of the relationship between the time of surgery and prognosis, the overall prognosis for patients who had surgery before 2010 and those who had surgery after 2010 was almost similar. Reports with a high proportion of stage I showed a good prognosis. The rate of total gastrectomy, the proportion of lymph node dissection above D1 + , or surgical complications had no effect on prognosis. Patients with gastric cancer aged 80 years or older who underwent radical surgery had a 5-year survival rate of up to 57%. Postoperative complications appeared to have a minor impact.
Keywords: Gastric cancer, Elderly, 80 years old
Introduction
The number of elderly patients with gastric cancer has increased in the past years as life expectancy has increased [1]. According to Japan’s national cancer registry data, the number of patients of gastric cancer is steadily increasing as the population of the elderly increases [2]. There have been many reports on elderly patients aged 75 and over in the past, but reports of patients with gastric cancer aged 80 and 85 and over have recently increased. Gastric cancer mortality has been decreasing due to improved treatment outcomes and an increase in patients with stage I gastric cancer. However, as the number of elderly patients affected was increasing, the number of deaths aged 80 years or older also increased. The number of deaths from gastric cancer in the 80 s was twice that of the 70 s and four times that of the 60 s.
In general, elderly patients experience many postoperative complications and have poor short- and long-term survival results. Hamilton et al. conducted a multicenter study with 3637 patients [3]. They found that, independent of preoperative comorbidity or health status, elderly patients above 75 years had an increased mortality rate after gastrectomy. However, the postoperative survival rate in patients with gastric cancer aged 80 years or older is unclear. Since half of the reports on elderly gastric cancer surgery patients were in Japanese, a meta-analysis of all publications, including the medical journal, is necessary. Other than PubMed, the Ichushi-Web database, which includes all publications in Japan, is one of the largest databases of medical research reports, containing over 13 million articles. In general, there is no meta-analysis of Japanese papers published by the Ichushi-Web database in PubMed. As of December 2021, there had been no meta-analysis of surgical cases of patients with gastric cancer aged 80 years or older, including articles published in medical journals.
Therefore, we systematically assessed the surgical cases of patients with gastric cancer for the elderly aged 80 years and over using a meta-analysis covering the literature published in PubMed and the Ichushi-Web database between 2010 and 2021.
Methods
Since surgical procedures, lymph node dissection, and chemotherapy vary depending on the historical background, we have limited the reports to those published after 2010. PubMed searched the literature published in the Ichushi-Web database using the keywords “Gastric cancer and elderly and 80 years old” and “Gastric cancer and 80 years old” from January 2010 to November 2021. A total of 253 PubMed reports and 366 Ichushi-Web database reports were extracted. Of these, 83 and 62 reports of gastric cancer over the age of 80 were reported, respectively (Fig. 1). Surgical treatment, 5-year survival rate, gender ratio, stage, surgical procedure, lymph node dissection, and postoperative complications were also considered. Of these papers, 10 literatures [4–13] were 80 years old or older with 50 patients or more (Fig. 1).
Fig. 1.
The selection flow chart for systematic review
Author, title, year of publication, journal published, and type of surgery were also retrieved and organized.
Analysis method
Five-year survival rates for each study were extracted. For studies that reported only 3-year survival rates, 5-year survival rates were calculated assuming that survival time followed an exponential distribution. Since no studies reported confidence intervals or standard errors, the standard errors were calculated assuming a binomial distribution for 5-year survival proportion. The average 5-year survival rate and the weights of the studies were calculated using random-effect models. We used a funnel plot and a forest plot to detect the publication bias. Meta-regression analysis was performed using the study features, including the proportion of studies published after 2020, stage I, total gastrectomy, lymph node resection, and complication, to investigate the source of prognosis variability. R package meta was used for the analysis. The size of the plotting symbol was proportional to the weight from the meta-analysis weight.
Results
Characteristics of the patients
The patients’ background of the selected reports is shown in Table 1. Of the ten reports identified, eight were from Japan [4–11] and two were from Taiwan [12] and South Korea [13]. The reporting year was from 2012 to 2020, and the study period for each report was from 1991 to 2015. The patients had to be 80 years or older or 85 years or older, and the number of patients ranged from 55 to 217. The female-to-male ratio was 1:1.74, 751 males and 431 females. Stage I accounted for 46% of the total, ranging from 23 to 59%.
Table 1.
Summary of the reports showing clinicopathological characteristics of elderly patients with gastric cancer
| Factor number of patients | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Author, Year, [Refs.] | Period | Age (years old) | Number of Patients | Male | Stage I | Total gastrectomy | Lymph node dissection by D1+/D2 | Complications | 5-year overall survival |
| Isobe et al. 2012, [4] | 1991 to 2010 | 85 ≦ | 217 | 139 | 115 | 45 | 43 | 59 | 73.8 |
| 64% | 53% | 21% | 20% | 27% | |||||
| Ohinata et al. 2017, [5] | 2007 to 2015 | 80 ≦ | 162 | 122 | 89 | 47 | 86 | 14 | 71.8 |
| 75% | 55% | 29% | 53% | 9% | |||||
| Terai et al. 2019, [6] | 1998 to 2015 | 80 ≦ | 130 | 76 | 52 | 36 | 82.6 | 46 | 45.2 |
| 58% | 40% | 28% | 64% | 35% | |||||
| Takahashi et al. 2020, [7] | 2009 to 2015 | 80 ≦ | 127 | 89 | 75 | 34 | 101 | 15 | 43.8 |
| 70% | 59% | 27% | 80% | 12% | |||||
| Sakurai et al. 2015, [8] | 2003 to 2010 | 80 ≦ | 95 | 55 | 43 | 28 | 95 | 23 | 43.2 |
| 58% | 45% | 29% | 100% | 24% | |||||
| Kiyokawa et al. 2015, [9] | 2000 to 2012 | 85 ≦ | 77 | 47 | 35 | 25 | 60 | 89 | 44.2 |
| 61% | 45% | 32% | 78% | 116% | |||||
| Endo et al. 2016, [10] | 1996 to 2015 | 85 ≦ | 56 | 26 | 29 | 0 | NA | 12 | 47.9 |
| 46% | 52% | 0% | 21% | ||||||
| Hikage et al. 2018, [11] | 2002 to 2015 | 85 ≦ | 55 | 29 | 28 | 13 | 44 | 25 | 63.6 |
| 53% | 51% | 24% | 80% | 45% | |||||
| Hsu et al 2012, [12] | 1994 to 2006 | 80 ≦ | 164 | 103 | 37 | 39 | 164 | 30 | 43.6 |
| 63% | 23% | 24% | 100% | 18% | |||||
| Kwon et al. 2014, [13] | 2005 to 2010 | 80 ≦ | 99 | 65 | 35 | 27 | 69 | 33 | 57.4 |
| 66% | 35% | 27% | 70% | 33% | |||||
The proportion of total gastrectomy was 24%, ranging from 21 to 32%, whereas the proportion of D1 + or D2 lymph node dissection was 72%, with a range of 20–100%. Postoperative complications occurred in 34% of cases, and surgical complications occurred in 16% of them. The 5-year survival rate accounted for 54%, with a range of 43–74%.
Forest plot and funnel plot of a 5-year survival rate in patients with gastric cancer over 80 years old.
The average 5-year survival rate and 95% confidence intervals from a meta-analysis are shown in Fig. 2. The weights calculated based on each trial’s standard error ranged from 4.3 to 21.6% (Fig. 2). The average 5-year survival rate was 57%. The survival rate was relatively good in two reports with a large number of patients, but it tended to be poor in five reports with a small number of patients, based on Japanese reports (Fig. 3). On the other hand, even in reports with a large number of patients, the survival rate in reports from overseas was not good. The survival outcomes of elderly patients who underwent gastric cancer surgery may vary substantially from report to report due to the large variability.
Fig. 2.
Forest plot of 5-year survival rate in gastric cancer patients over 80 years old. The standard error of individual study was calculated based on binomial approximation. The weights were calculated using a random effect model
Fig. 3.

Estimated standard error plotted against 5-year survival rate. Survival rates in the two studies with small standard errors (Isobe et al. [4], Oohinata et al. [5]) were better than in the other studies. This trend was opposite to the results that could occur with publication bias
Comparison with 5-year survival rate
Meta-regression with the proportion of patients after 2010 as exploratory variable is shown in Fig. 4. Despite an increase in the number of patients after 2010, the survival rate did not improve considerably. Figure 5 shows the relationship between the stage I proportion and the survival rate. With a higher proportion of patients with stage I gastric cancer in the facility, the survival rate was significantly higher.
Fig. 4.

We analyzed whether the overall prognosis was improved by the proportion of patients after 2010. Even though the proportion of patients after 2010 increased, the overall prognosis did not improve significantly
Fig. 5.

The relationship between the stage I proportion and the survival rate was shown. As a matter of course, the survival rate was significantly higher when the proportion of stage I in the facility was high, and the survival rate was significantly lower when the proportion of stage II or higher was high
The relationship between the proportion of total gastrectomy and survival rate in the surgical procedure for gastric cancer is shown in Fig. 6. As the proportion of total gastrectomy increased, the overall survival did not worsen considerably. Lymph node dissection is evaluated by dividing it into a group of D0, D1, and D1 + or higher. Figure 7 shows the relationship between the proportion of D1 + or higher and the survival rate. The higher the proportion of D1 + or higher, the lower the survival rate. Figure 8 shows the relationship between the rate of all complications and the rate of survival. Increasing the incidence of complications did not appear to have a substantial impact on survival.
Fig. 6.

Regarding the surgical procedure for gastric cancer, the relationship between the proportion of total gastrectomy and the survival rate is shown. Survival did not significantly worsen as the rate of total gastrectomy increased
Fig. 7.

Lymph node dissection is evaluated by dividing it into a group of D0 and D1 and a group of D1 + or higher, and the relationship between the proportion of D1 + or higher and the survival rate is shown. The higher the proportion of D1 + or higher, the lower the survival rate
Fig. 8.

The relationship between the rate of all complications and the survival rate is shown. Increasing the incidence of complications did not significantly reduce survival
Discussion
We conducted a meta-analysis of gastric cancer over the age of 80 from 2010 to 2021 using ten reports for which sufficient data could be extracted from PubMed and the Ichushi-Web database. Despite the fact that the proportion of surgical cases after 2010 was high, there was no substantial change in the survival rate. Furthermore, there was little association between the frequency of postoperative complications and survival rate.
Although the proportion of patients after 2010 was high, the survival rate did not improve. There are two plausible causes for this: (i) Older patients (85 + or 90 years) may undergo surgery after 2010 and (ii) older people may have more comorbidities and higher postoperative complications [4, 14]. Nashimoto et al. [15] showed that 6 ~ 8% of gastric cancer patients were aged over than 80-years old between 2002 and 2009. On the other hand, 15% (3685 of 24,508) of gastric cancer patients were aged over than 80-years old in 2013 based on national registration (Japanese Gastric Cancer Association) [https://www.jgca.jp/entry.html]. Generally, elderly age was independently associated with increased morbidity after GC resection [3]. Hamilton et al. concluded that increased risk in elderly gastric cancer patient should be recognized and considered in indications for resection.
Patients who underwent total gastrectomy and those who underwent distal gastrectomy had the same survival rate. The following two points are presumed to be the reasons: (i) The proportion of patients who had total gastrectomy was almost the same among institutions, making it difficult to analyze this tendency, and (ii) the indication for patients with total resection was thought to be accurate. The difference between the frequency of postoperative complications and the survival rate was small because postoperative management proficiency and the indication for surgery were both appropriate.
Even in the elderly, the survival rate improved as the proportion of patients with stage I gastric cancer increased, suggesting that diagnosing and treating patients with stage I gastric cancer at an early stage had merit. Regarding the relationship between lymph node dissection and survival rate, the higher the frequency of standard dissection corresponding to stage II or III, the lower the survival rate. This could be attributed to a reduced overall survival rate when the proportion of patients with advanced cancer increases. Furthermore, more frequent standard lymph node dissections in the elderly may result in partial over-invasion.
The slope of the graph shown in Fig. 5 was 0.4439. The higher the degree of progression, the wider the extent of lymph node dissection; therefore, the ratios other than stage I and the ratios of D1 + /D2 dissection are considered almost the same. The slope of the graph seen in Fig. 7 was almost the same as 0.4069, implying that the proportion of patients with D1 + /D2 dissection was similar to the proportion of patients other than stage I. Even at advanced stages, the extent of dissection may have been limited in certain patients due to age and risk.
There was little association between the frequency of postoperative complication and survival rate. Such tendency was controversial to previous reports on non-elderly patients [16]. Since the description of complications is different for each report in our current, only total frequency of all complications could be extracted. Therefore, the unclear frequency of serious complications that affect prognosis may be the reason for the lack of relevance. Furthermore, although infectious complications such as pneumonia have been reported to be associated with prognosis, it is possible that other non-infectious complications are not associated with prognosis [17, 18]. In this report, there are few reports that mention the type and severity of complications in detail, so there is a possibility that they are biased.
This study has several limitations: (i) The papers analyzed in this meta-analysis were reports of gastric cancer surgery in patients 80 years old and above. However, since age and gender were not clearly defined, the age distribution and the male-to-female ratio may differ, which may affect the variability in 5-year survival rates. (ii) Since there is no clear explanation of prognosis-related factors such as body mass index (BMI) and serum albumin levels in the target paper, it was not possible to analyze the effect on prognosis. Yamakawa et al.’s report [19] showed that preoperative BMI and serum albumin levels were significantly lower in the group aged 85 years and older. So, actual raw data will be required in future multi-institutional studies. (iii) In terms of the lymph node dissection extent, it is very likely that age-related reduction surgery was performed. However, since there was no specific description of the actual extent of lymph node dissection, the relationship with prognosis could not be analyzed. (iv) Since several papers did not describe the 5-year survival rate (reporting only the 3-year survival rate) and all papers did not report the standard error or confidence interval for the survival rate, the estimated values assuming a statistical distribution were used. (v) In terms of adjuvant chemotherapy for elderly patients, most recent treatment guideline does not show specific statements. Recent reports showed that perioperative treatment compliance, especially in the postoperative phase, was poorer in elderly patients compared with non-elderly patients. Neo-adjuvant treatment may be the other option for elderly patients with advanced stages [20, 21].
In conclusion, patients aged 80 years and older who underwent gastric cancer surgery had a good 5-year survival rate of 57%, and postoperative complications appeared to have a minor impact. However, since the report lacked detailed data on individual patient age or comorbidity, multi-institutional studies based on individual raw data seem to be necessary. Moreover, since there is little data on patients over 90 years old, it is important to collect and analyze real-world data including this age group in the future.
Author contributions
HS designed the study. CO analyzed the data. HS, MO and YO supervised the study. NN and HS wrote the manuscript.
Declarations
Conflict of interest
The authors have no conflict of interest to declare.
Ethical approval
Ethical approval is not required because all data in this study used Ichushi-Web database.
Informed consent
This article does not contain any study with humans or animals as participants that is performed by any of the authors. Informed consent is not required because all data in this study used Ichushi-Web database.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Endo S, Fujiwara Y, Higashida M, et al. Survival analyses of elderly gastric cancer patients with or without surgery. Surg Today. 2021 doi: 10.1007/s00595-021-02303-8. [DOI] [PubMed] [Google Scholar]
- 2.Trends in site-specific crude mortality rate 1965–2014. Cancer Statistics in Japan-2018, Tokyo, Japan, 2018; pp 40. https:// ganjoho.jp/data/reg_stat/statistics/brochure/2018/cancer_statistics_2018.pdf. Accessed Jan 2022
- 3.Hamilton TD, Mahar AL, Haas B, et al. The impact of advanced age on short-term outcomes following gastric cancer resection: an ACS-NSQIP analysis. Gastric Cancer. 2018;21:710–719. doi: 10.1007/s10120-017-0786-6. [DOI] [PubMed] [Google Scholar]
- 4.Isobe T, Hashimoto K, Kizaki J, et al. Surgical procedure, complication, and prognosis for gastric cancer in the very elderly (<85): a retrospective study. Kurume Med J. 2012;59:61–70. doi: 10.2739/kurumemedj.59.61. [DOI] [PubMed] [Google Scholar]
- 5.Oohinata R, Iwasaki Y. Gastrectomy for gastric cancer in elderly patients (in Japanese) Clin Gastroenterol. 2017;32:689–697. [Google Scholar]
- 6.Terai T, Watanabe A, Kuroda Y, et al. Evaluation of surgical treatment for patients with gastric cancer aged over 80 years (in Japanese) J Nara Pref Gen Med C. 2019;23:23–27. [Google Scholar]
- 7.Takahashi T, Kaneoka Y, Maeda A, et al. The preoperative prognostic nutrition index is a prognostic indicator for survival in elderly gastric cancer patients after gastrectomy: a propensity score-matched analysis. Updates Surgery. 2020;72:483–491. doi: 10.1007/s13304-020-00745-2. [DOI] [PubMed] [Google Scholar]
- 8.Sakurai K, Muguruma K, Nagahara H, et al. The outcome of surgical treatment for elderly patients with gastric carcinoma. J Surg Oncol. 2015;111:848–854. doi: 10.1002/jso.23896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kiyokawa T, Hiki N, Nunobe S, et al. Feasibility of gastrectomy with standard lymphadenectomy for patients over 85 years old with gastric cancer. Ann Surg Oncol. 2015;22:3962–3969. doi: 10.1245/s10434-015-4489-0. [DOI] [PubMed] [Google Scholar]
- 10.Endo S, Shimizu Y, Ikenaga M, et al. Survival benefit of gastrectomy for gastric cancers in patients ≧85 years old: a retrospective propensity score-matched analysis. Surgery. 2017;161:984–994. doi: 10.1016/j.surg.2016.10.012. [DOI] [PubMed] [Google Scholar]
- 11.Hikage M, Tokunaga M, Makuuchi R, et al. Surgical outcomes after gastrectomy in very elderly patients with gastric cancer. Surg Today. 2018;48:773–782. doi: 10.1007/s00595-018-1651-x. [DOI] [PubMed] [Google Scholar]
- 12.Hsu JT, Liu MS, Wang F, et al. Standard radical gastrectomy in octogenarians and nonagenarians with gastric cancer: are short-term surgical results and long-term survival substantial? J Gastrointest Surg. 2012;16:728–737. doi: 10.1007/s11605-012-1835-4. [DOI] [PubMed] [Google Scholar]
- 13.Kwon IG, Cho I, Guner A, et al. Minimally invasive surgery as a treatment option for gastric cancer in the elderly: comparison with open surgery for patients 80 years and older. Surg Endosc. 2015;29:2321–2330. doi: 10.1007/s00464-014-3955-2. [DOI] [PubMed] [Google Scholar]
- 14.Komori K, Kano K, Aoyama T, et al. The short- and long-term outcomes of gastrectomy in elderly patients with gastric cancer. In Vivo. 2020;34(5):2697–2703. doi: 10.21873/invivo.12090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Nashimoto A, Akazawa K, Isobe Y, et al. Gastric cancer treated in 2002 in Japan: 2009 annual report of the JGCA nationwide registry. Gastric Cancer. 2013;16:1–27. doi: 10.1007/s10120-012-0163-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Shimada H, Fukagawa T, Haga Y, et al. Does postoperative morbidity worsen the oncological outcome after radical surgery for gastrointestinal cancers? A systematic review of the literature. Ann Gastroenterol Surg. 2017;1:11–23. doi: 10.1002/ags3.12002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Cho H, Tsuchida K, Iwasaki K, et al. Risk factors of post-operative pneumonia in elderly patients with gastric cancer: a retrospective cohort study. Jpn J Clin Oncol. 2021;51(7):1044–1050. doi: 10.1093/jjco/hyab032. [DOI] [PubMed] [Google Scholar]
- 18.Endo S, Fujiwara Y, Higashida M, et al. Survival analyses of elderly gastric cancer patients with or without surgery. Surg Today. 2022;52:75–83. doi: 10.1007/s00595-021-02303-8. [DOI] [PubMed] [Google Scholar]
- 19.Yamakawa Y, Bando E, Kawamura T, et al. A clinical study of surgical resection for patients over 85 years of age with gastric cancer (in Japanese) Jpn J Gastroenterol Surg. 2014;47:1–10. doi: 10.5833/jjgs.2013.0011. [DOI] [Google Scholar]
- 20.Slagter AE, Tudelaet B, van Amelsfoort RM, et al. Older versus younger adults with gastric cancer receiving perioperative treatment: results from the CRITICS trial. Eur J Cancer. 2020;130:146–154. doi: 10.1016/j.ejca.2020.02.008. [DOI] [PubMed] [Google Scholar]
- 21.Shen L, Zhang X, Kong L, et al. The clinical features, management, and survival of elderly patients with gastric cancer. J Gastrointest Oncol. 2022;13(2):605–614. doi: 10.21037/jgo-22-280. [DOI] [PMC free article] [PubMed] [Google Scholar]


