Prognostic Model for One-year Mortality in Patients with Upper Gastrointestinal Malignancy at Cipto Mangunkusumo Hospital Jakarta, Indonesia

Khaira Utia Yusrie1,2, Dadang Makmun1,2, Andri Sanityoso Sulaiman2, Siti Setiati2

1 Division of Gastroenterology. Department of Internal Medicine, Faculty of Medicine University of Indonesia;
2 Cipto Mangunkusumo National General Hospital Jakarta, Indoneia.

Conflict-of-interest statement: The author(s) declare(s) that there is no conflict of interest regarding the publication of this paper.

Open-Access: This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http: //creativecommons.org/licenses/by-nc/4.0/

Correspondence to: Khaira Utia Yusrie, Division of Gastroenterology. Department of Internal Medicine, Faculty of Medicine University of Indonesia.
Email: utiyusrie@yahoo.com

Received: Jauuary 24, 2022
Revised: March 8, 2022
Accepted: March 11, 2022
Published online: April 21, 2022


Aims: To identify prognostic factors for one-year mortality in patients with upper gastrointestinal malignancy in Indonesia.

Methods: Retrospective cohort study using the hospital database of patients with upper gastrointestinal malignancy at Cipto Mangunkusumo Hospital (2015-2019). Bivariate and multivariate cox proportional hazards regression analysis were performed to identify independent factors associated with mortality upper gastrointestinal malignancy. Scoring system were developed based on the identified factors.

Results: 184 patients were analyzed, mostly male (58.7%) with average ages 54.5 years old. Independent factors associated with one-year mortality were age > 60 years with HR 1,93 (95%CI 1.30; 2.88), body mass index < 20 with HR 2.04 (95%CI 1.25; 3.33), smoking history with HR 1.77 (95%CI 1.20; 2.61), performance status ECOG > 2 with HR 3.37 (95%CI 2.11; 5.37), clinical stage which is 4th stage HR 9.42 (95%CI 1.27; 69.98) and 3rd stage HR 9.78 (95%CI 1.31; 72.69), and cellular differentiation grade with HR 2.30 (95%CI 1.48; 3.58). 1-year survival rate was 39,7% with median survival was 9 months. The scoring system for predicting mortality had AUC values of 0,918 respectively.

Conclusion: The independent factors associated with one-year mortality were age, body mass index, smoking history, performance status, clinical stage of the tumour, and cellular differentiation grade. 1-year survival rate was 39.7%. The mortality probability prediction scoring system has been developed for upper gastrointestinal malignancy.

Key words: Upper gastrointestinal malignancy; Mortality prognostic; Mortality probability prediction score

© 2022 The Author(s). Published by ACT Publishing Group Ltd. All rights reserved.

Yusrie KU, Makmun D, Sulaiman AS, Setiati S. Prognostic Model for One-year Mortality in Patients with Upper Gastrointestinal Malignancy at Cipto Mangunkusumo Hospital Jakarta, Indonesia. Journal of Gastroenterology and Hepatology Research 2022; 11(2): 3706-3712 Available from: URL: http: //www.ghrnet.org/index.php/joghr/article/view/3253


Upper gastrointestinal malignancy especially esophageal and gastric cancer is the sixth and third leading cause of cancer-related deaths worldwide. Its mortality number has gradually decreased because of the increasing capability in diagnostic early stage of the cancer. The decrease of mortality has been profounded in developed countries such as Japan, Korea, European Countries, Ukraine, Columbia and Ecuador[1]. However, in under developed countries, the mortality rate caused by upper gastrointestinal malignancy especially esophageal and gastric cancer have not yet been profounded. As in the case of Indonesia, the number of survivors of Upper gastrointestinal malignancy especially esophageal, gastric and duodenum has unfortunately not yet been concluded.

As depicted in the study of the longterm surviving probability in China for the period of year 2000 until 2011, it has been concluded that the number of malignancy for male patient has been relatively stable, however, it is has been found a prolific increasing rate for female patients. As supported by the study, there have been four major malignancies types found; they are lungs, gastric, kidney and esophageal cancer. These malignancy types are accounted for 57% of the entire malignancies occurred in China as opposed to around 18% occurred in US[1]. In Indonesia, A study conducted by Makmun D. et al, has found a changing prevelancy trend taking place during two different periods: the period of 2002-2006 and 2007-2011. The study compares the changes of malignancy rate in each different cases during those two periods. The colorectal malignancy case found was about 76.3% during the former period and then decreased during the latter period, and followed by the gaster of malignancy case at or around 15.6% and 14.9%, Esophageal cancer case at or around 7.4% and 7.6% respectively, and the last is duodenal cancer case which took a significant increase from 0.7% to 6.1% respectively from the entire cases of the upper gastrointestinal malignancy. Prevalency of esophageal adenocarsinoma case has increased siginificantly from 36.4% to 69.2%. The average age of the patients experiencing the upper gastrointestinal malignancy has also decreased; from 53.02 ± 13.12 to 50.43 ± 11.93 (p = 0.031), and the prevelancy rate has markedly improved from 60.9% to 82.4% (p = 0.018), however, for the age ≥ 60, the case has significantly decreased from 34.8% to 13.7% (p = 0.036)[2].

Based on the study in China, it is found that the number of deaths caused by esophageal cancer is about 3.940 of 6.493 cases (or around 60.7%) and caused by gastric cancer patients is around 5.228 of 8.537 cases (or around 61.2%) with surviving median around 2.99 years for esophageal cancer and 2.92 years for gastric cancer. The cohort retrospective study conducted by Charati et al, in Iran during 2006-2013 has found that the median surviving years for gastric cancer is around 19 ± 2.04 months with the surviving rate of the patients during 1st year until 5th year outlined as follows: 64%, 44%, 34%, 24% and 19% respectively[3].

Detecting the cases in early stage has been a virtue in treatment or conducting prognostic the patients. In the US study, some 67% of patients were detected at Stage III and IV compared to only 10% found at Stage I. Correlation the case with lymph has also given varied surviving rate. The Surveillance, Epidemiology and Results Programs (SEER) has collected data in the period of 1973-2000 research suggesting that there has been a correlation with the amount of lymph involved and also served as independent prognosis factor in overall survival[4,5].

To date, a fair number of researches have been conducted and these researches have analyzed some factors that contribute to the surviving rate of the patient with gastrointestinal malignancy: i.e.: esophageal cancer, gastric cancer and duodenal cancer. These researches had been conducted with all of data limitation confronting the respective reserachers. Tustumi et al has conducted research prognostic factors on esophageal cancer and reported that there has been a correlation with age, gender, weight loss, body mass index, performance status, and tumour stage taken as the influential factors in detoriation of prognosis of the patients[6]. The study conducted by Samadi et.all on patients with esophageal and gastric cancer has suggested that smoking and the degree of tumour differentiation are the primary factors impacting the patient’s surviving rate[7]. Other study conducted by Jiang S. et al on Duodenum malignancy has suggested that the patients > 60 years of age with tumour case would lead to higher hazard ratio i.e. age between 59-75 would result in 1.59 (95%CI 1.2-2.1) with p = 0.001. While those older than 75 would result in 2.74 (95%CI 2.0-3.7) with p < 0.001[8]. Maezawa et al also conducted their cohort retrospective research during the year of 2000 until 2015 and concluded that cormobidity with the highest CCI scoring is the independent risk factor on deterioration prognosis with gaster cancer[9].

Based on the reseach on relevant literatures, we have not yet concured a study that purposedly combines the entire prognostic factors leading to mortality on patients with upper gastrointestinal malignancy. Furthermore, there has not yet been available to us a scoring system that could be applied to asess mortality prognostic on the patients with gastrointestinal malignancy so that it enables the clinics to evaluate and to conduct the proper intervention required.


The prognostic study with cohort restrospective design has been conducted which based on the medical record provided by Cipto Mangunkusumo Hospital Jakarta during the period of August-December 2020. The patients as samples being mínimum 18 years of age with gastrointestinal malignancy case including esophageal, gastric and duodenal cancer that had been recorded for the period of 2015-2019.

Sampling methodology is conducted by tracing and evaluating medical records of the patients of Cipto Mangunkusumo Hospital. The method applied is total sampling with consecutive basis so that it would suffice to attain minimum data sample calculation. The collected data encompass the characteristics of research study including age, gender, smoking history, type of malignancy as evidenced by supportive tools (i.e. histopatology, radiology figures and endoscopy results), accompanying illness for cormobid, nutrient status with body mass index, history of loss weight, and stage tumour. In the event that patient last state / condition is not traceable or not conclusive because of the lost of follow-up, therefore, the last treatment date in the medical record is utilized as a proxy the liveliness of the patient and this would be applied in the “Censor” and accounted for in the surviving rate calculation.

The entire data collected would be recorded in the table and loaded in the Micsrosoft Excel softaware, and processed further by utilizing a program “Stata 15” in the computer. Basic chararcteristics and clinic of the research subject would be presented in a table.

Each predictor variable would be analyzed with bivariate and multivariate methods with statistical test Cox Proportional Hazards Regression Model. In bivariate analysis, each predictor would become covariant to the mortality in 1 year, so that Hazard Ratio (HR) could be calculated along with each respective p value. Predictor factor with p value < 0.25 would be added in the multivariate analysis by applying Cox Regression Model technique simultaneously so that Hazard Ratio could be retrived as well as the p value for each predictor, where the value would be decided as significant (meaningful) to the occurance of mortality if and only if the p < 0.05. The significance of the variables are the independent factor to the mortality of the patients with gastrointestinal malignancy.

In attaining the prediction of mortality of the patients with gastrointestinal malignancy, this research would be developed with scoring system from multivariate analysis. The entrie variables with p < 0.05 woud be analysed and subsequently included in the Scoring system stage. The score is developed by combining coefficient value (B) from the variables produced by Cox Regression Model, then evaluating the lowest score, followed by calculating score. In deciding the crossing value from logistic scoring to assess mortality prediction, the curve of Receiver Operating Characteristics (ROC) would be analyzed so as to gain the sensitivity values and specificities concerning to the performing predictor model. Performing score would be evaluated by the Area Under Cure (AUC). The ROC value is deemed as acceptable if it results AUC > 0.8.

To evaluate the capability of generalisation from this prognostic model, internal validaton would be applied to the existing data.

Mortality data woud also be undergone surviving analysis through Kaplan-Meier Curve to achieve the cumulated surviving and the entire surviving of the subject for 1 year of the observed period., as well as surviving of the subject for each related variable. To attest statistical significance, it is decided if each of the variable results in p < 0.05 hence categorized as significant.

This research has been recognized and passed the ethical conduct test as issued by The Commitee of the Ethics Medical Research of Medical Faculty of University of Indonesia., as evidenceded with refference letter numbered: KET-1402/ UN2.F1/ETIK/PPM.00.02/2020.


For the obeservation period of January 2015 until December 2019, there had been 184 patients were analysed in this research. 108 of whom were male (58.75%) and 76 remaining were female (41.3%). The average age of female was 54.5 years of age. The largest proportion of gastrointestinal malignancy was found to be gastric cases with 84 cases (45.7%) and followed by esophageal cancer with 69 cases (37.5%), and finally, duodenal cancer with 31 cases (16.8%) as depicted in table 1. The proportion of gastrointestinal malignancy accounts for 60,3% with the surviving median rate is around 9 months (Figure 1).

Figure 1 One-year survival Kaplan-Meier Curve.

Table 1 Characteristics of the data.
VariableMeasurement (n = 184)
Age> 60 yo64 (34.8%)
< 60 yo120 (65.2%)
SexMale108 (58.7%)
Female76 (41.3%)
BMI< 20104 (56.5%)
> 2080 (43.5%)
Performance statusECOG > 271 (38.6%)
ECOG < 2113 (61.4%)
Weight loss> 5 kg110 (59.8%)
< 5 kg74 (40.2%)
SmokingSmoking80 (43.5%)
Not smoking104 (56.5%)
Tumour stageStage 488 (47.8%)
Stage 372 (39.1%)
Stage 0-224 (13.0%)
GradingPoorly and undifferentiated95 (51.6%)
Well and moderately differentiated89 (48.4%)
ComorbidCCI score > 473 (39.7%)
CCI score <4111 (60.3%)
LocationDuodenal31 (16.8%)
Gaster84 (45.7%)
Esophagus69 (37.5%)

Bivariate and Multivariate Analysis

In order to ascertain the prognostic factors that are impacted to the mortality of the patients with gastrointestinal malignancy, bivariate and multivariates analysis were applied.

The assessment as guided under the proportional hazard assumption is carried out to attest whether each predictive variable would conform or fulfil to the requirement of Cox’s assumption before further analysis of bivariate and multivariate taking place.

In bivariate analysis with the following characteristics: age (p = 0.009), body mass index (p < 0.001), weight loss (p < 0.001), smoking (p = 0.002), tumour stage (p = 0.001), the grade of cellular differentiation (p < 0.001), cormobid (p < 0.001), these are the prognostic factors that correlate with the death caused by upper gastrointestinal malignancy. There had not yet been found a significant correlation between gender and location of the tumour carried by the patients with the mortality prognostic of upper gastrointestinal malignancy (Table 2).

Based on the result of multivariate analysis as the final model of prognostic independent factors. The patients with upper gastrointestinal malignancy would be depicted in table 3.

Table 2 The result of bivariate analysis of prognostic factors related with mortality of patients with upper gastrointestinal malignancy.
VariableMeasurementnEventHR min CI max CI p-value
Age> 60 yo64471.651.132.410.009
< 60 yo12064
BMI< 20104793.342.25.06<0.001
> 208032
Performance statusECOG > 271666.724.5110.03<0.001
ECOG < 211345
Weight loss> 5 kg110823.322.165.09<0.001
< 5 kg7429
Not smoking10452
Tumour stageStage 4886732.044.44231.090.001
Stage 3724319.292.65140.160.003
Stage 0-2241
GradingPoorly and undifferentiated95804.032.656.14<0.001
Well and moderately differentiated8931
Comorbid CCI score > 473592.691.843.92<0.001
CCI score <411152

Table 3 Multivariate analysis.
VariableHR (95%CI)p-value
Age1.93 (1.30-2.88)0.001
BMI2.04 (1.25-3.33)0.004
Smoking1.77 (1.20-2.61)0.004
Performance status3.37 (2.11-5.37))<0.001
Stage 49.42 (1.27-69.98)0.028
Stage 39.78 (1.31-72.69)0.026
Grading2.30 (1.48-3.58)<0.001

Scoring System in Mortality prediction with upper gastrointestinal malignancy

This study has developed a scoring system for asessment the prognostic mortality of the patients with upper gastrointestinal malignancy. The entire variables in this analysis in p value < 0.05 to be eligible for the next step that is the scoring system development. The scoring system is developed by calculating the coeficient B and standard error (SE). Table 4 belows has shown the related steps in developing scoring system.

Table 4 Scoring System.
      Step 1Step 2Step 3
Age (>60 yo)0.660.23.251.491
BMI (<20)0.710.252.841.31
Performance status ( >2)
Grading (poorly and undifferentiated)0.830.223.711.692

The entire related steps taken have provided scores for each respective variable as depicted in table 5, where the score of 10 being the maximum limit and ROC curve is applied to evelalute the scoring system shown in figure 2. The figure 2 shows the Area Under Curve (AUC) of 0.918 (95%CI 0.88-0.96) with intersection point ≥ 4 (Sensitivity 0.88; Speficity 0.73).

Table 5 Variable Scoring.
Age> 60 yo1
< 60 yo0
Body Mass Index< 201
> 200
Performance statusECOG > 22
ECOG < 20
StageStage 3-41
Stage 0-20
Cellular differentiationPoorly and undiffenrentiated2
Well and moderately differentiated
Figure 2 ROC Curve.

Based on the evaluation of scoring system on predicting factors for patients survival with upper gastrointestinal malignancy, the validity of the predicting survival factors on patients with gastrointestinal malignancy has been found to be acceptable as evidenced in AUC score of 0.918 (95%CI 0.88-0.96).

The scoring system had been developed and had been evaluated on its internal validity by going through 10 times validation check on 20% of the subjects of the research, selected randomly. On each validation test, the AUC values had been retrieved which fall around >0.8 with p < 0.001.

Development of the mortality prediction model with scoring system

The mortality prediction model had been developed with a scoring system so as to enable to calculate predicted probability of patient mortality during a certain period, the model applied is as follows: 1-S0(t)^(exp(y)) with S0(t) = baseline survival in certain t and y = 0,55*total score. The application of the prediction of mortality probability in a a certain period is described in table 6.

Table 6 Prediction of mortality probability in a certain period.
   Mortality prediction
Month!st month2nd month3rd month4th month5th month6th month7th month8th month9th month10th month11th month
ScoreBaseline survival0.9540.9130.8630.8020.6880.6250.5830.510.4910.4220.321
Probability formula : 1-S0(t)^(exp(y)) y = 0.550.


This research essentially involves the cohort restrospective study on 184 patients with upper gastrointestinal malignancy at Cipto Mangunkusumo Hospital. The subject of the study comprises male (58.7%) being the majority sample, age < 60 years (62.5%) with median age stood at or around 54.5 years of age. This study reveals the general figure of the patients who experience upper gastrointestinal malignancy where the majority cases fall into Gastric cancer (45.7%) which is coincidently similar with that of China. The China’s study found that 53.7% esophageal cancer occurred in male whereas 77.8% cases fell into gastric cancer[1,7]. However, in the Iran’s study, the result is different with that of Indonesia where majority cases of upper gastrointestinal malignancy experienced by patients’ age < 60 years, on the other side, Iran’s study shows different cases where upper gastrointestinal malignancy experienced by patients’ age > 60 years accounting for around 68.8% with esophageal cancer cases and for around 73.3% with gastric cancer cases. Other characteristic shows similarity between Indonesia and Iran studies i.e. upper gastrointestinal malignancy experienced by Indonesia’s patients who do not smoke accounted for a relatively higher at or around 56.5%, this result is relatively similar with Iran’s case which falls around 56.5% for esophageal cancer and around 64.5% of gastric cancer[7].

This study finds that the frequency of tumour location for the patients with upper gastrointestinal malignancy occurred in the patients with gastric cancer case is the highest - which stood around 45.7%- among other malignancies types. This case is similar with that of Iran where gastric cancer occurance stood at 66,4% compared to esophageal cancer at around 34.6%[7].


This study also finds that a number of factors including age, body mass index, smoking, performance status, and cellular differentiation grade is considered to be the factors that independently contribute to the mortality of the patients with upper gastrointestinal malignancy. This study also suggest that patient with age ≥ 60 years is exposed to a higher probability (likelihood) at or around 1.93 times than those < 60 years of age. This finding also concurs with other study by Guzman et al, with HR 2.9 times (95%CI 1.3-6.8) and furthermore, it is also concurred with study by Maezawa et al where the HR stood at or around 4.46 (95%CI 1.07-2.95), as is the case with the study by Jiang S. et al for the study period of 2004-2014 resulting HR value around 1.59 (CI 95% 1.2-2.1) found in the patients age bracket 59-75 years and HR 2.741 (95%CI 2.0-3.7)[9,10,11].

Body mass index contributes to HR value of 2.04 (95%CI 1.25-3.33) as prognostic mortality for the patients with upper gastrointestinal malignancy. This HR value is a little higher than that of Maezawa et al study where their HR finding stood around 1.61 (95%CI 1.41-2.26) for gastric cancer, BMI value from this study also a little lower than that of Tustumi et al study for the patients with adenocarsiom esophagus whose HR value stood at or around 4,42 (95%CI 1.09;1.79)[6,9].

Smoking contributes to HR value at 1.77(95%CI 1.21-2.67) concluded from the subject study for the likelyhood of mortality caused by upper gastrointestinal malignancy. The number does not markedly differ with that of Samadi et al study where the HR value was around 1,8 (95%CI 1.023-3.23) and that of Smyth et al with HR value of 1.48[7,12].

Performance status contributes to HR value around 3.37 with 95%CI 2.11-5.37 where in this factor, the patients with ECOG ≥ 2 are exposed to a higher degree around 3.37 times higher for mortality to occur. This HR value is relatively lower than that of Maezawa et al study whose HR value stood at around 1.51 (95%CI 1.23-1.86)[9].

Tumour stage also contributes to a higher HR value, where in the patients with Stage 3 results in HR value 9.78 (95%CI 1.31-72.69), those with stage 4 would result HR at 9.42 (95% 1.27-69.98). These numbers are relatively higher than those of Jian et al study where the HR value received at or around 3.69 (95%CI 2.0-6.7) for those paients with stage 3 and HR 6.48 (95%CI 2.0-6.7) for those with stage 4[11].

The grade of tumour differentiation contributes to HR value of 2.30 (95%CI 1.48-3.58) for the probability of mortality occurance caused by upper gastrointestinal malignancy. This number does not markedly differ with that of Tustumi et al study whose HR value stood around 2.0 (95%CI 1.1-3.65) related to the case of esophageal cancer[6].

This research also describes a general relationship between late intervention and the hazard ratio that may occur in order to improve the scoring performance of the patients so that mortality rate would eventually decrease. The study of Faber J. et al on the patients with esophageal cancer and esophagogastric junction suggests that nutrient intervention may improve weights, performance status and PGE2 level of the patients with esophageal cancer before undergoing Chemotherapy[13].

This research has also found that majority of the patients were found to already be at last stage which accounts for around 47.8% with stage 4 and 39.1% with stage 3, so that a number of proper interventions had been carefully evaluated to prolong the patients life expectancy or surviving rate. The analysis study conducted by Ge XY et al proposes that based on SEER data analysis, the patients with gaster cáncer at Stage 4 who had undergone intervention and chemotherapy had a better surviving rate compared to have been left untreated. Their analysis suggests that the surviving rate of the patients undergone intervention had lived longer for about 16 months as compared to 4 months for those who had not, where the entire factors of the patients prognostics had been taken into consideration. Other study suggests that the surviving rate is correlated with the nutrient status, age and existing cormobid[14,15].

The study conducted by Silvers MA et al proposes that nutritional intervention before conducting invervention therapy would result in lower mortality for the patients with upper gastrointestinal malignancy, although the sample taken in this study is relatively small; around 21 patients taken randomly, so that it is imperative to conduct a study with rather larger samples to realize significant meaning the importance of the study[16].

The scoring system that we had developed does also require certain cautions to be exercised especially in interpretating the result or outcome produced by the scoring system. The inherent limitation are relatively contributed by the characteristic of the sample patients who were mostly categorized as advance stage, therefore, it is imperative to conduct further research on the patients with varied stage.

One-year surival

In this study, surviving rate during 1 year period for the patients with upper gastrointestinal malignancy results in around 39.7%, with mortality rate of 60.3%. The mortality rate is similar with that of Hua Z et al study in December 2015, in which the mortality of esophageal cancer stood at 60.7% and of gastric cancer 61.2%[1]. The Hua findings are relatively higher than those of Samadi et al in their 2007’s study, which finds the surviving rate during 1 year is around 40.5% and surviving during 2 years is around 16.9%, nevertheless, median surviving rate is not markedly different for 9 months from Samadi et al compared to 7 months from the other study in Pakistan[7]. Decreasing mortality number has been more evident in developed countries which has better early detection system as the case in Japan, Korea and European countries. One study concludes that the decreasing rate of mortality in a number of countries have mostly correlated with the much ealier detection and better health management for those cases in earlier stage and by applying new techniques.

Scoring sytem of predicting mortality in upper gastrointestinal malignancy

As high as 60.3% mortality proportion has been received from this research, therefore, it is important to conduct an evaluation that could differentiate between the patients with higher risk and those with lower risk during the illness of the patient. By doing so, the risks could be assessed and some proper intervention actions could be taken for the patients with upper gastrointestinal malignancy. This study has also been able to identify prognostic factors for patients with upper gastrointestinal malignancy and eventualy utilized as a base to develop a scoring system that could predict mortality in certain period of time. Scoring system developed in this study applies a number of clinical variables that had been designed to be inputted by the clinics with ease. Such variables include age, body mass index, smoking record, performance status with related ECOG score, tumour stage, and the cellular differentiation grade. The scoring system resulted from this study shows rather better performance whose Area Under Curve (AUC) result is around 0.918 (95%CI 0.88-0.96).

To our best knowledge, this study presents latest data about mortality and mortality prognostic model for the patients with upper gastrointestinal malignancy in Indonesia and a number of variables have also been included in this study which are relatively easy to be evaluated by the medics in secondary level as well as tersier level. The data of the variables in evaluation would easly be received and considered as objective so that the application in different or varied clinical condition can be executed. The development of prognostic model with scoring system in this research could also help the clinics in predicting mortality risk of the patients with upper gastrointestinal malignancy.

This research has been of cohort retropective research and has been based upon the relevant medical record available, therefore, certain miss-information or bias may occur. The subject of the resarch largely involve with patients with advance stage, so that it is imperative to exercise proper and correct evaluation in relation to postpone or extend the medical intervetion required on a patient to improve prognosis of the patient. The scoring performance developed in this study does not include some specific variable which is important in evaluating or assessing the prognostic mortality factor for patient with upper gastrointestinal malignancy. The reasoning is that some specific markers had not yet been exercised in regular basis on the patient with upper gastrointestinal malignancy. We also suggest that external validation shall be exercised before the mortality scoring system of upper gastrointestinal malignancy would be applied in the clinic.


Age, body mass index, smoking, performance status, tumour stage, and cellular differentiation grade are the contributing factors to mortality of the patient with upper gastrointestinal malignancy during first year at Cipto Mangunkusumo Hospital. The surviving rate in first 1 year for the patient with upper gastrointestinal malignancy in Cipto Mangunkusumo Hospital has stood at or around 39.7% with surviving median of 9 months. The scoring system could be developed for daily clinic application to initiate mortality prediction of the patient with upper gastrointestinal malignancy based on the result of better AUC’s score and it had undergone internal validation. It is imperative to initiate external validation upon this prognostic mortality scoring system, since this study was developed with prospective design.


1. Hua Z, Zheng X, Xue H, Wang J, Yao J. Long-term trends and survival analysis of esophageal and gastric cancer in Yangzhong, 1991-2013. PLos ONE 12(3): e0173896 [internet]. [cited 2020 Jul 12]. [DOI: 10.1371/journal.pone.0173896]; [PMID: 28288195]

2. Makmun D, Simadibrata M, Abdullah M, Syam AF, Fauzi A, Renaldi K, Rani A, Krisnuhoni E. Changing trends in gastrointestinal malignancy in Indonesia: The Jakarta experience. J Cancer Res Ther. 2014; 2(9): 160-8. [DOI: 10.14312/2052-4994.2014-24] Corpus ID 55059667

3. Charati JY, Janbabaei G, Alipour N, Mohammad S, Gholiabad SG, Fendereski A. Survival prediction of gastric cancer patients by Artificial Neural Network model. Gastroenterol Hepatol Bend Bench 2018; 11: 110-7 [PMID: 29910851]

4. Elimova E, Kaya DM, Harada K, Ajani JA. Potentially curable cancers of esofagus and stomach. Mayo Clin Proc 2016; 91(9): 1307-18 [DOI: 10.1016/j.mayocp.2016.07.018]; [PMID: 27594190]

5. Japanese Gastric Cancer Association. Japanese gastric cancer treatment guideline 2018. Gastric Cancer 2020 [DOI: 10.1007/s10120-020-01042-y]; [PMID: 32060757]

6. Tustumi F, Kimura CMS, Takeda FR, Uema RH, Salum RAA, Riberiro-junior U, Cecconello I. Prognostic factors and survival analysis in esophageal carcinoma. ABCD Arq Bras Cir Dig 2016. 29 (3): 138-41 [DOI: 10.1590/0102-6720201600030003]; [PMID: 27759773]

7. Samadi F, babaei M, Yazdanbod A, Fallah M, Nouraie M, Nasrollahzaedah D, Sadjadi A, Derakhsan MH, Shokuhi B, Fuladi R, Malekzadeh R. Survival rate of gastric and esophageal cancers in Ardabil province, nort-west of Iran. Arch Iranian Med 2007; 10(1): 32-7 [PMID: 17198451]

8. Jiang QL, Huang XH, Chen YT, Zhang JW, Wang CF. Prognostic factors and clinical characteristics of patients with primary duodenal adenocarcinoma: A single-center experience from China. Biomed research International 2016. [internet]. [cited 2020 Jul 12]. [DOI: /10.1155/2016/6491049] Article ID 6491049

9. Maezawa Y, Aoyama T, Kano K, Tamagawa H, Numata M, Hara K, Murakawa M, Yamada T, SatoT, Ogata T, OshimaT, Yukawa N, Yoshikawa T, Masuda M, Rino Y. Impact of the age-adjusted Charlson comorbidity index on the short-and long term outcomes of patients undergoing curative gastrectomy for gastric cancer. Journal of Cancer 2019; 10: 5527-35 [DOI: 10.7150/jca.35465]; [PMID: 31632496]

10. Guzman RF, Mercado DV, Castillo ML, Rios LEG, Rodriguez DG. Prognostic factors for survival in patients with resectable advanced gastric adenocarcinoma. Cirugia Y Cirujanos 2016; 84: 469-76 [DOI: 10/1016/j.circir.2016.01.007]; [PMID 27039288]

11. Jiang S, Zhao R, Li Y, Han X, Liu Z, Ge W, Dong Y, Han W . Prognosis and normogram for predicting postoperative survival of duodenal adenocarcinoma: A retrospective study in China and the SEER database. Scientific report 2018; 8: 7940 [DOI: 10.1038/s41598-018-26145-6]; [PMID: 29786691]

12. Smyth EC, Capanu M, Janjigian YY, Kelsen DK, Colt D, Strong VE, Shah MA. Tobacco use is associated with increased recurrence and death from gastric cancer. Ann Surg Oncol 2012: 19 (7): 2088-94 [DOI: 10.1245/s10434-012-2230-9]; [PMID: 22395977]

13. Faber J, Uitdehaag MJ, Spaander M, Steenbergen-Langeveld S, Vos P, Berkhout M, Lamers C, Rumke H, Tilanus H, Siersema P, van Helvoort A, vander Gaast A. Improved body weight and performance status and reduced serum PGE2 levels after nutritional intervention with a specific medical food in newly diagnosed patents with esophageal cancer or adenocarcinoma of the gastro-esophageal junction. Journal of Cachexia, Sarcopenia and Muscle 2015; 6: 32-4 [DOI: 10.1002/jcsm.12009]; [PMID 26136410]

14. Ge XY, Ge F, Wang Z, Wang YL, Lei LW, Liu QR, Sun XY, Jiang X. Analysis of risk factors of stage IV gastric cancer from the SEER database. Ann R Coll Surg Engl 2020; 00: 1-8 [DOI: 10.1308/rcsann.2020.0031]; [PMID 32326735]

15. Yamaguchi K, Yoshida K, Tanahashi T, Takahashi T, Matsuhashi N, Tanaka Y, Tanabe K, Ohdan H. The long term survival of stage IV gastric cancer patients with conversion therapy. Gastric Cancer 2018; 21: 315-23 [DOI: 10.1007/s10120-017-0738-1]; [PMID: 28616743]

16. Silvers MA, Savva J, Huggins CE, TRuby H, Haines T. Potential benefits of early nutritional intervention in adults with upper gastrointestinal cancer : a pilot randomised trial. Support Care Cancer 2014; 22: 3035-44 [DOI: 10.1007/s00520-014-2311-3]; [PMID: 24908429]


  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.