Open Access

Prediction of survival benefit when deciding between chemotherapy and best supportive therapy in elderly patients with advanced gastric cancer: A retrospective cohort study

  • Authors:
    • Aya Sugimoto
    • Tsutomu Nishida
    • Naoto Osugi
    • Kei Takahashi
    • Kaori Mukai
    • Dai Nakamatsu
    • Tokuhiro Matsubara
    • Shiro Hayashi
    • Masashi Yamamoto
    • Sachiko Nakajima
    • Koji Fukui
    • Masami Inada
  • View Affiliations

  • Published online on: November 19, 2018     https://doi.org/10.3892/mco.2018.1772
  • Pages: 83-91
  • Copyright: © Sugimoto et al. This is an open access article distributed under the terms of Creative Commons Attribution License.

Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )
Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )


Abstract

It remains unclear whether elderly patients (EPs) with advanced gastric cancer (AGC) benefit from chemotherapy. The aim of the present study was to examine the prognostic factors for EPs with AGC in order to generate a prognosis‑predicting scoring system. This single‑center retrospective study examined consecutive patients with AGC between April 2012 and July 2017. Risk factors for survival in EPs aged ≥75 years were identified using a Cox proportional hazards model, and a prognostic scoring system was generated and retrospectively evaluated to determine its usefulness for predicting patient prognosis. A total of 61 patients were enrolled as EPs (mean age, 81 years) and compared with 80 non‑EPs (mean age, 66 years). The median survival time (MST) was significantly longer for non‑EPs compared with that for EPs (3.8 vs. 10.1 months, respectively; P=0.0447). Among the EPs, 29 (48%) received chemotherapy and 32 received best supportive care (BSC). A total of 68 non‑EPs (85%) received chemotherapy and 12 non‑EPs received BSC. Among EPs with AGC, age‑adjusted multivariate analysis revealed that performance status (PS), neutrophil/lymphocyte ratio (NLR)<4, intestinal‑type histology and chemotherapy were significant prognostic factors. To predict EPs too frail for chemotherapy prior to treatment, one point was assigned for a PS of 1, diffuse‑type histology and NLR≥4, whereas 2 points were assigned for PS≥2, and the point totals for each patient were calculated. A cut‑off point of 2 had the best P‑value by the log‑rank test and was used to divide the patients into low‑risk (LoR: Score 0‑1) and high‑risk (HiR: Score 2‑4) groups. The MST of the LoR and HiR groups was 23.6 and 3.6 months, respectively (P<0.001). As regards treatment strategies and risk groups, the LoR chemotherapy group had the best prognosis (P=0.0010), and LoR EPs who were administered chemotherapy had a longer MST (30.3 months) compared with EPs who received BSC (8.7 months). In conclusion, scoring systems using PS, histology and NLR may be useful when considering chemotherapy in EPs with AGC.

Introduction

Gastric cancer is the fifth most common type of cancer (1) and the third most common cause of cancer-related mortality worldwide (2). The majority of patients with gastric cancer are diagnosed between their late 60s and 80s. Consequently, the number of elderly patients (EPs) with advanced gastric cancer (AGC) has increased as a result of population aging and prolonged life expectancy; therefore, most AGC patients are elderly. Regarding treatment benefit for patients with AGC, a meta-analysis of overall survival (OS) in 35 clinical trials including a total of 5,726 patients revealed that chemotherapy significantly improved survival compared with best supportive care (BSC) [hazard ratio (HR)=0.37; 95% confidence interval (CI): 0.24–0.55] (3). However, EPs (aged ≥75 years) often have multiple chronic medical conditions and develop adverse events induced by chemotherapy, as these individuals are frailer compared with younger patients with AGC. Consequently, EPs are usually excluded from the indications or are not considered to be suitable for chemotherapy to narrow the target when evaluating the efficacy and adverse events of chemotherapy in most clinical trials for AGC. Therefore, in practice, there is little evidence on whether EPs with AGC can tolerate chemotherapy and whether chemotherapy is beneficial in terms of survival.

The chemotherapy indications for EPs with AGC primarily depend on performance status (PS) and comorbidities. The attending physician subjectively decides upon the treatment strategy (chemotherapy vs. BSC) based on patient information rather than objective criteria. Therefore, it is difficult to distinguish between fit and frail EPs prior to treatment. The host's systemic inflammatory response (SIR) has recently emerged as a critical factor based on objective information on tumor progression (4). SIR includes common inflammation-based prognostic scores, such as C-reactive protein (CRP) (5); the Glasgow Prognostic Score (GPS) (6), which includes CRP and albumin; and the neutrophil/lymphocyte ratio (NLR) (7). Previous studies have revealed that SIR is associated with poor outcome in several types of cancer (8). However, whether SIR is useful for distinguishing between fit and frail EPs for chemotherapy remains unclear. Therefore, we hypothesized that scoring based on a combination of several SIR markers and other background characteristics on initial evaluation may predict the lifetime survival benefit for EPs with AGC, and help identify EPs who are likely to benefit from chemotherapy. In the present study, prognostic factors for EPs with AGC were examined to generate a prognostic scoring system to predict EPs who may be too frail for chemotherapy prior to treatment, and a subset from this scoring system was retrospectively evaluated to help predict prognosis or decide on treatment strategies.

Patients and methods

Patients

The present study was a single-center retrospective study including consecutive patients with AGC at Toyonaka Municipal Hospital (Toyonaka, Japan) between April 2012 and July 2017. During the study period, 173 patients diagnosed with AGC without surgical indication visited our department. Patients with surgically resectable disease were excluded and all patients enrolled had clinically confirmed evidence of metastatic lesions or unresectable factors. Patients who received chemotherapy were evaluated for the best response according to the Response Evaluation Criteria in Solid Tumors (version 1.1) and treatment toxicity according to the National Cancer Institute Common Terminology Criteria for Adverse Events (version 3.0). The present study was conducted in accordance with the principles outlined in the Declaration of Helsinki, and approval was obtained from the Institutional Review Board. This was a retrospective study involving human data that were previously collected and did not require the additional recruitment of human subjects; thus, the need for informed consent was waived via the opt-out method on our hospital website.

Treatment strategy and follow-up

The treatment strategy (chemotherapy or BSC) was decided by the attending physician according to the patient's background. The patients were routinely followed-up, including physical examination, laboratory testing, dynamic computed tomography scanning and esophagogastroduodenoscopy. If BSC was decided for a patient on initial evaluation, their physician was subsequently contacted to request follow-up information or investigate census registries. The latest follow-up was in August 2017. OS was calculated from the date of initial clinical diagnosis of AGC until death or the last available follow-up prior to death from any cause. Surviving patients were censored on their last follow-up date.

Data collection

The following factors were collected from medical records at the time of AGC diagnosis: Patient demographics [age, sex, Eastern Cooperative Oncology Group (ECOG) PS]; primary tumor locations; indications for chemotherapy; histological type; serum albumin level; CRP; NLR; estimated glomerular filtration rate (eGFR); tumor markers [serum carcinoembryonic antigen (CEA) and carbohydrate antigen (CA) 19–9]; number of metastatic organs, including lymph nodes; and Charlson Comorbidity Index (9). Clinical course was also investigated, including chemotherapy agent and mean relative dose intensity (RDI), best response under chemotherapy, subsequent therapies and survival status. The RDI was defined as the ratio of the delivered dose intensity of each chemotherapy drug to the standard (referenced) dose intensity in the regimen. In the present study, RDI was evaluated during the planned first course of the first-line chemotherapy. The mean RDI was calculated as the mean percentage of RDI for all drugs in the first-line regimen.

Statistical analysis and prognostic scoring system

Median and range are reported for continuous variables. Categorical variables are summarized as frequency (percentage). Differences in variables were evaluated using the Wilcoxon signed-rank test or χ2 test. OS was estimated using the Kaplan-Meier method and compared using the log-rank test. HR and CI were estimated for OS.

Prognostic factors for survival in EPs were examined at the time of diagnosis of AGC via univariate and multivariate analyses using a Cox proportional hazards model. Subsequently, significant prognostic factors were identified and then clinical thresholds were set based on the best log-rank P-value. One point was assigned for each significant factor and the sum of the total points was calculated. Next, the best cut-off point to predict prognosis was evaluated. Finally, a prognostic scoring system was generated and the cut-off point was set.

All reported P-values are two-sided, and P<0.05 is considered to indicate statistically significant differences. Statistical analyses were performed using JMP statistical software (version 13.1. 0, SAS Institute Inc., Cary, NC, USA).

Results

Patient characteristics

Of 173 patients with AGC, 143 had clinically confirmed evidence of unresectability or recurrence. Ultimately, 141 patients were enrolled in the present study, as 2 patients were treated at another hospital after diagnosis. The patient characteristics are summarized in Table I. S-1 plus cisplatin has been recommended as a standard treatment for AGC in Japan (10) based on the SPIRITS trial (11). The SPIRITS trial, however, was conducted with patients aged <75 years. For that reason, to evaluate the benefit of chemotherapy in EPs with AGC aged ≥75 years, they were compared with non-EPs aged <75 years during the same study period. A total of 61 EPs [39 men; mean age, 81 years (range, 75–93 years)] and 80 non-EPs [59 men; mean age, 66 years (range, 36–74 years)] as a control group were enrolled in the present study. Of those patients, 29 EPs (48%) and 68 non-EPs (85%) received chemotherapy, whereas 32 EPs and 12 non-EPs received BSC based on the judgement of their attending physician. The reasons for BSC included the patient's wish (n=12), poor PS (n=9), advanced age (n=6), dementia (n=2) and others (n=3) for Eps, and poor PS (n=9), patient's wish (n=1) and others (n=2) for non-EPs.

Table I.

Characteristics of elderly patients and non-elderly patients with advanced gastric cancer.

Table I.

Characteristics of elderly patients and non-elderly patients with advanced gastric cancer.

Elderly patients, total no.Non-elderly patients, total no.


CharacteristicsAll, 61CT, 29BSC, 32P-valueAll, 80CT, 68BSC, 12P-value
Male sex, no. (%)39 (64)21 (72)18 (56)0.189259 (74)49 (72)10 (83)0.4131
Median age, years (range)81 (75–93)8085.50.000366 (36–74)66680.2271
Median BMI, kg/m2 (range)20.3 (14–32)21.220.20.5203  20.5 (14.2–30.5)20.323.40.1179
Diabetes mellitus, no. (%)10 (16)2 (6.9)8 (25)0.056515 (19)13 (19)2 (17)0.8410
Hypertension, no. (%)34 (56)19 (66)15 (47)0.143222 (28)17 (25)5 (42)0.2332
Dementia, no. (%)6 (9.8)0 (0)6 (19)0.01413 (3.8)2 (2.9)1 (8.3)0.3647
Chronic kidney disease, no. (%)48 (79)21 (72)27 (84)0.254616 (20)15 (22)1 (8.3)0.2731
Charlson Comorbidity Indexa0010.02180010.1156
Low/medium/high + very highb37/22/222/7/015/15/20.047352/24/4/047/18/3/05/6/1/00.1845
PS 0–1, no. (%)46 (75)26 (90020 (63)0.032463 (79)62 (91)1 (8.3)<0.0001
Histology type, intestinal, no. (%)25 (41)14 (48)11 (34)0.270325 (31)21 (31)4 (33)0.8659
Gastric cardia tumor site, no. (%)12 (20)8 (28)4 (13)0.229617 (21)16 (24)1 (8.3)0.2355
Target lesion, no. (%)53 (87)26 (90)27 (84)0.541873 (91)65 (96)8 (67)0.00118
No. of metastatic lesions ≤1, no. (%)25 (41)14 (48)11 (34)0.270327 (34)21 (31)6 (50)0.1966
CEA, median (range)7.711.26.90.26924.45.12.90.4846
(1–7446) (0.7–1737)
CA19-9, median (range)221095.50.069121.524150.4952
(2–205307) (2–267426)
Albumin ≥3 g/dl, no. (%)34 (56)21 (72)13 (41)0.012656 (74)52 (79)4 (40)0.0094
CRP level <1 mg/dl, no. (%)31 (52)18 (62)13 (42)0.118938 (49)37 (55)1 (9.1)0.0046
NLR<4, no. (%)29 (48)17 (61)12 (38)0.072640 (52)36 (55)4 (36)0.2638

a The Charlson Comorbidity index encompasses 19 medical conditions weighted 1–6 with total scores ranging from 0 to 37.

b For each condition, the scores reflected the risk as follows: 0, low; 1–2, medium; 3–4, high; and ≥5, very high risk. CT, chemotherapy; BSC, best supportive care; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; NLR, neutrophil/lymphocyte ratio; PS, performance status; CEA, carcinoembryonic antigen; CA19-9, cancer antigen 19-9. Chronic kidney disease was defined as an eGFR <60 ml/min.

A total of 53 EPs (87%) and 73 non-EPs (91%) had a target lesion. The median follow-up period was 187 days (range, 8–2,728 days). During the follow-up period, there were a total of 119 deaths. The causes of death included cancer [n=50 (chemotherapy, n=23 and BSC: n=27)] and treatment-related complications (n=1) for EPs, and cancer [n=65 (chemotherapy, n=55 and BSC, n=10), treatment-related complications (n=2) and other causes [n=1 (chemotherapy, n=1 and BSC, n=0)] for non-EPs.

OS

The OS data for EPs and non-EPs are plotted in Fig. 1A. The median survival time (MST) of non-EPs was significantly longer compared with that of EPs (10.1 vs. 3.8 months, respectively; P=0.0447); however, longer survival curves appeared to overlap 2 years after the initial diagnosis (Fig. 1A).

The survival curves grouped by age, with or without chemotherapy, are shown in Fig. 1B. There was a highly significant difference in OS between chemotherapy and BSC in non-EPs (10.8 vs. 1.6 months, respectively; P<0.0001), as well as in EPs (6.6 vs. 2.8 months, respectively; P=0.0069). There was, however, a smaller lifetime survival benefit in EPs compared with that in non-EPs.

Efficacy, toxicity and feasibility of chemotherapy in EPs with AGC

The details of the chemotherapy regimen are shown in Table II. Among EPs who were administered chemotherapy, the response rate (RR) was 17%, which was similar to that of non-EPs (21%). The disease control rate (DCR) was lower in EPs compared with that in non-EPs (51 vs. 71%, respectively), but this difference was not statistically significant. The median average RDI in EPs was 90% (mean average RDI: 80%). Among EPs treated with chemotherapy, second-line treatment was administered to 7 patients, which was a significantly lower percentage compared with that among non-EPs (24 vs. 68%, respectively; P<0.0001).

Table II.

Chemotherapy regimens and responses of patients with advanced gastric cancer.

Table II.

Chemotherapy regimens and responses of patients with advanced gastric cancer.

ItemsEPs, n=29Non-EPs, n=68P-value
First-line chemotherapy regimen, n (%)
  S-1 ± trastuzumab16 (53)5 (7.4)
  S-1 + CDDP ± trastuzumab11 (37)41 (60)
  S-1 + CDDP + docetaxel0 (0)2 (2.9)
  S-1 + L-OHP ± trastuzumab or ramucirumab1 (3.3)5 (7.4)
  CAP + trastuzumab0 (0)1 (1.5)
  CAP + CDDP ± trastuzumab1 (3.3)4 (5.9)
  CAP + CPT-11 + trastuzumab0 (0)1 (1.5)
  Paclitaxel or nab-paclitaxel1 (3.3)9 (13)
Second-line chemotherapy, n (%)7 (24)46 (68)<0.0001
Treatment-related deaths, n (%)1 (3.3)2 (2.9)1.000
Response to first-line chemotherapy, n (%)
  CR0 (0)0 (0)0.2464
  PR5 (17)14 (21)
  SDa10 (34)34 (50)
  PD7 (24)7 (10)
  NE7 (24)13 (19)
  RR5 (17)14 (21)0.7866
  DCR, n (%)15 (51)48 (71)0.1035
    Median average RDI, n (%)901000.0165

a Including non-CR/non-PD in case of patients without target lesions. CDDP, cisplatin; CAP, capecitabine; L-OHP, oxaliplatin; CPT-11, irinotecan; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; NE, not evaluated; RR, response rate; DCR, disease control rate; RDI, relative dose intensity.

Among patients who received chemotherapy, all toxicity grades were observed in 22 patients (81%), and grade >3 toxicity was observed in 6 patients (22%) (Table III). Of those patients, 2 continued the same regimen after dose reduction, but the remaining patients discontinued the first chemotherapy treatment as a result of grade 4 neutropenia or grade 3 fatigue, acneiform rash, palmar-plantar erythrodysesthesia syndrome or nausea. Finally, 6 patients (22%) discontinued the first chemotherapy treatment as a result of treatment failure; 1 patient succumbed to febrile neutropenia.

Table III.

Toxicities in elderly patients with advanced gastric cancer receiving chemotherapy.

Table III.

Toxicities in elderly patients with advanced gastric cancer receiving chemotherapy.

ToxicitiesAll gradesGrade 3 or 4Grade 5
Hematological, n (%)
  White blood cell decrease5 (17)3 (10)0
  Neutropenia4 (14)3 (10)0
  Anemia2 (7.0)00
  Thrombocytopenia1 (3.4)00
  Febrile neutropenia1 (3.4)01 (3.4)
Non-hematological, n (%)
  Malaise12 (41)1 (3.4)0
  Nausea9 (31)1 (3.4)0
  Diarrhea4 (14)00
  Acute kidney injury1 (3.4)00
  Palmar-plantar erythrodysesthesia syndrome1 (3.4)1 (3.4)0
  Peripheral sensory neuropathy1 (3.4)00
  Acneiform rash3 (10)1 (3.4)0
  Skin hyperpigmentation1 (3.4)00
  Acute coronary syndrome1 (3.4)00
Prognostic factors and scoring system in EPs with AGC

The mean CRP level in EPs was 2.5±3.4 mg/dl, with a median of 0.69 mg/dl (range, 0–13.7 mg/dl). The log-rank P-values calculated for the CRP level were P=0.3053 (<0.5 vs. ≥0.5 mg/dl), P=0.3014 (<1 vs. ≥1 mg/dl) and P=0.9347 (<1.5 vs. ≥1.5 mg/dl). The mean albumin level in EPs was 3.0±0.6 g/dl, with a median of 3 g/dl (range, 1.5–4.1 g/dl). The log-rank P-values calculated for albumin were P=0.5548 (<2.5 vs. ≥2.5 g/dl), P=0.040 (<3 vs. ≥3 g/dl) and P=0.5662 (<3.5 vs. ≥3.5 g/dl). The mean NLR in EPs was 7.4±8.5, with a median of 4.3 (range, 1.2–46.5). The log-rank P-values calculated for NLR were P=0.2447 (<3.5 vs. ≥3.5), P=0.0374 (<4 vs. ≥4) and P=0.1586 (<4.5 vs. ≥4.5). Based on the cut-off P-values for CRP, albumin and NLR, optimal cut-off levels of 1 mg/dl, 3.0 g/dl and 4, respectively, were determined. Subsequently, a univariate analysis was performed to identify significant prognostic factors. The results revealed that the following factors were significantly associated with better prognosis (Table IV): PS 0 or 1, intestinal-type histology, albumin level ≥3 g/dl, CRP level <1 mg/dl, NLR<4, number of metastatic lesions ≤1 and treatment strategy. The age-adjusted multivariate analysis using these 7 significant factors revealed that PS 0 (HR=0.27, 95% CI: 0.10–0.75, P=0.0113), PS 1 (HR=0.42, 95% CI: 0.17–0.797, P=0.0274), intestinal-type histology (HR=0.41, 95% CI: 0.22–0.77, P=0.0053), NLR<4 (HR=0.35, 95% CI: 0.14–0.79, P=0.0109) and treatment strategy (HR=0.42, 95% CI: 0.19–0.92, P=0.0313) were significant prognostic factors (Table IV). To predict EPs too frail for chemotherapy prior to treatment, 1 point was assigned for PS of 1, diffuse-type histology and NLR≥4, and 2 points for PS≥2, except for treatment strategy (Table V), and the sum of the total points for each patient was calculated as the score. Subsequently, 6 patients (10%) with a total score of 0, 10 patients (17%) with a score of 1, 22 patients (37%) with a score of 2, 15 patients (25%) with a score of 3 and 4 patients (12%) with a score of 4 were identified (Table IV). The survival curves for each score were well-separated (Fig. 2A). The mean total score among EPs was 2.1±1.1 (median, 2). The log-rank P-values calculated for the total score were P=0.0016 (0 vs. ≥1, HR=0.21), P<0.0001 (0–1 vs. 2–4, HR=0.21) and P<0.0001 (0–2 vs. 3–4, HR=0.23). Therefore, a cut-off point of 2 was set (total score 0–1 vs. 2–4). EPs with a total score of 2–4 were classified as the high-risk (HiR) group (n=44; chemotherapy, n=17 and BSC, n=27), and the remaining patients were classified as the low-risk (LoR) group (n=16; chemotherapy, n=11 and BSC, n=5). Among EPs, a score of 0–1 was a significantly good predictor of OS (age-adjusted HR for a score of 2–4: 0.21, 95% CI: 0.09–0.44; P<0.0001). MST was significantly longer in the LoR compared with that in the HiR group (23.6 vs. 3.6 months, respectively; P<0.0001). Among the 4 groups based on treatment strategy and risk, the data revealed that LoR-chemotherapy had the best prognosis (P=0.0010) and that MST was longer in EPs treated with chemotherapy (n=11; MST, 30.3 months) compared with those treated with BSC in the LoR group (n=5; MST, 8.7 months). The HiR-chemotherapy group had a significantly longer MST compared with the HiR-BSC group (4.2 vs. 2.4 months, respectively; P=0.0322; Fig. 2B).

Table IV.

Prognostic factors for overall survival in elderly patients with advanced gastric cancer based on the univariate analysis and on the multivariate analysis adjusted by age.

Table IV.

Prognostic factors for overall survival in elderly patients with advanced gastric cancer based on the univariate analysis and on the multivariate analysis adjusted by age.

Univariate analysisMultivariate analysis adjusted by age


FactorsHR95% CIP-valueHR95% CIP-value
Age, years
  <800.650.34–1.180.1588
  ≥801
Sex
  Male0.800.44–1.490.4667
  Female1
Performance status
  00.130.05–0.32<0.00010.270.10–0.750.0113
  10.370.15–0.620.00150.420.17–0.970.0274
  ≥21 1
CCI
  High/very high1
  Medium1.000.20–18.20.9967
  Low0.980.21–17.50.9819
Histology
  Intestinal type0.560.31–0.990.04900.410.22–0.770.0053
  Diffuse or other1 1
Albumin level, g/dl
  ≥30.380.21–0.680.00121.000.45–2.200.9942
  <31 1
CRP level, mg/dl
  <10.280.15–0.52<0.00010.480.20–1.120.0917
  ≥11 1
eGFR
  <600.830.44–1.710.5918
  ≥601
NLR
  <40.250.13–0.48<0.00010.350.14–0.790.0109
  ≥41 1
Metastatic lesions
  0–10.470.125–0.850.01140.650.31–1.300.2222
  ≥21 1
Treatment strategy
  Chemotherapy0.470.26–0.820.00860.420.19–0.920.0313
  BSC1 1

[i] CI, confidence interval; HR, hazard ratio; BSC, best supportive care; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; NLR, neutrophil/lymphocyte ratio; CCI, Charlson Comorbidity Index.

Table V.

Scoring indices for predicting survival in elderly patients with advanced gastric cancer.

Table V.

Scoring indices for predicting survival in elderly patients with advanced gastric cancer.

Scoring systemScore
Performance status
  00
  11
  ≥22
Histology type
  Intestinal0
  Diffuse1
NLR
  <40
  ≥41
Total score (0–4)EPs, n (%)Non-EPs, n (%)
  06 (10)2 (2.6)
  110 (17)18 (23)
  222 (37)27 (35)
  315 (25)21 (27)
  47 (12)9 (12)
Data missing13
Total score 0–1, low risk16 (27)20 (26)
Total score 2–4, high risk44 (77)57 (74)

[i] EPs, elderly patients; non-EPs, non-elderly patients; NLR, neutrophil/lymphocyte ratio.

Similarly, non-EPs with a total score of 2–4 were classified as the HiR group (n=57; chemotherapy, n=46 and BSC, n=11), and the remaining patients were classified as the LoR group (n=20; chemotherapy, n=20 and BSC, n=0). The MST was significantly longer in the LoR compared with that in the HiR group (7.6 vs. 21.4 months, respectively; P=0.0016), which was similar to the results for non-EPs. In contrast to EPs, the HiR-chemotherapy non-EPs had a significantly longer MST compared with the HiR-BSC non-EPs (8.1 vs. 1.7 months, respectively; P<0.0001).

Discussion

The populations of industrial countries, including Japan, are aging rapidly. Although >60% of patients with newly developed cancer and >70% of cancer fatalities are reported in elderly populations in the United States and Europe (12), EPs are generally under-represented in clinical trials (13,14). Two retrospective analyses of clinical trials involving patients with solid tumors suggested that older age does not necessarily diminish the tolerance or response to chemotherapy (15); however, there is little evidence on the benefit of chemotherapy for EPs with cancer.

The majority of patients with gastric cancer are diagnosed between their late 60s and 80s. The SPIRITS trial showed an advantage for S-1, an oral FU antitumor drug, combined with cisplatin for the first-line treatment of AGC (11). Consequently, S-1 plus cisplatin was recommended as a standard treatment for AGC in Japan (10), although S-1 plus cisplatin was not found to be superior to 5-FU plus cisplatin in a global randomized trial conducted outside Japan (FLAGS) (16). The SPIRITS trial, however, included patients aged <75 years. In previous phase II studies, Koizumi et al (17) and Imamura et al (18) demonstrated that S-1 monotherapy was safe and efficacious for EPs with AGC. These two trials were conducted with 33 and 35 AGC patients, respectively, aged >75 years, and they reported RRs of 21 and 14%, respectively, and a median OS of 15.7 and 14.6 months, respectively. Regarding adverse events, anemia and anorexia were predominantly observed, but no treatment-related deaths occurred. In the SPIRITS trial, the median OS in younger patients receiving S-1 monotherapy was 11.0 months, which was shorter compared with that in the abovementioned studies in EPs. This finding reflects the enrollment of only selected EPs who were considered fit for chemotherapy in both studies.

Several recent studies have demonstrated that SIR is associated with prognosis in various types of cancer (5,6,8). Previous studies have reported that albumin, CRP and NLR are important prognostic factors in patients with gastric cancer and other types of cancer (1921). In the present study, we proposed a scoring system using 3 factors, namely PS, histological type and NLR, identified as significant in a multivariate analysis including SIR. This scoring system separated EPs who may benefit more from chemotherapy from those who may benefit less. In brief, MST was longer in EPs treated with chemotherapy in the LoR compared with those treated with chemotherapy in the HiR group. However, a clinically relevant chemotherapy benefit was not observed for EPs in the HiR group, as survival with chemotherapy was only prolonged from 2.4 to 4.2 months (Fig. 2B). Based on these results, this scoring system may be valuable for selecting EPs fit for chemotherapy, and risk factors should be taken into consideration when determining the indications for chemotherapy for EPs with AGC.

Several studies have reported the clinicopathological characteristics of EPs with gastric cancer. Arai et al (22), reported that differentiated-type adenocarcinoma accounted for 50% of advanced cancers among 994 Japanese patients aged ≥65 years. The proportion of differentiated-type carcinoma increased significantly with advancing age in female patients with advanced cancer, but no significant change was observed in male patients with advanced cancer. Other reports (23,24) have also found an increased proportion of differentiated-type carcinoma with increasing age. Mizoe et al (25), reported higher 5- and 10-year survival rates in patients with differentiated-type gastric cancer compared with patients with the undifferentiated type; the authors hypothesized that this difference may be due to the rapid tumor growth of undifferentiated-type cancer.

The present study has several limitations due to its retrospective nature. First, the number of patients at a single institute was small, and the patient sample was divided into two groups using 75 years as the cut-off age, as most clinical trials have been conducted with patients aged <75 years. Consequently, the age for non-EPs ranged from 36 to 74 years, with a wider distribution compared with that for EPs. Second, the treatment strategy or regimen relied upon the physician's decision; however, a strength of the present study was the evaluation of prognostic factors in patients, including non-EPs or patients receiving BSC with poor PS, compared with those receiving chemotherapy in the real world, rather than in a clinical trial, during the same period in the same hospital. Moreover, the present study identified potential EPs who may benefit from chemotherapy. Third, a common cut-off value for SIR has not been established. Several cut-off values are used for CRP and NLR, and these values are not consistent, even for the same type of cancer (26,27). In the present study, optimal cut-off levels were set for CRP (1 mg/dl), albumin (3.0 g/dl) and NLR (4). Zhang et al (19), evaluated whether NLR was a useful prognostic factor in patients with gastric cancer using a meta-analysis; the results of a subgroup analysis of previous studies demonstrated that the prognostic value of NLR was not substantially affected by the cut-off value. In the present study, that NLR was found to be of significant prognostic value in EPs. It may also be useful to consider NLR when determining indications for chemotherapy in frail patients with AGC. Finally, the screening tools for EPs were not sufficiently evaluated, although no significant factors were found based on the Charlson Comorbidity Index. The optimal screening tools for EPs have not yet been established. However, Decoster et al (28) conducted a systematic review based on 17 different screening tools in EPs, and concluded that G8 was the best screening tool due to its high sensitivity with acceptable specificity and prognostic value for predicting outcome measures. In the present study, the G8 health status scores were not routinely evaluated during the study period. In the future, however, the scoring system must be verified in a validation cohort compared with G8. Additionally, we did not evaluate quality of life (QOL) in this elderly population. Although chemotherapy is often used to improve QOL for patients with end-stage cancer, slightly extending the MST with chemotherapy must be weighed against QOL in HiR EPs.

In conclusion, the lifetime survival benefit from chemotherapy was found to be greater in non-EPs compared with that in EPs with AGC. However, EPs with AGC who were placed in the LoR group by the scoring system had reasonable OS compared with non-EPs with AGC. Therefore, EPs with AGC in the HiR group should be evaluated individually and receive special attention in order to determine the indications for chemotherapy.

Acknowledgements

Not applicable.

Funding

No funding was received.

Availability of data and materials

The datasets used during the present study are available from the corresponding author upon reasonable request.

Authors' contributions

AS and TN collected and analyzed the data and wrote the manuscript. AS, TN, NO, KT, KM, DN, TM, SH, MY, SN, KF and MI provided medical care for the patients enrolled in this study.

Ethics approval and consent to participate

The present study was conducted in accordance with the Declaration of Helsinki, and approval was obtained from the Institutional Review Board of the Toyonaka Municipal Hospital (2015-10-06).

Patient consent for publication

This is a retrospective study involving human data that was previously collected and did not require the additional recruitment of human subjects; thus, the need for informed consent was waived via the opt-out method of our hospital website.

Competing interests

The authors declare that they have no competing interests.

References

1 

International Agency for Research on Cancer: Estimated number of incident cases, both sexes, worldwide (top 10 cancer sites) in 2012. 2012.simplehttp://gco.iarc.fr/today/online-analysis-multi-bars?mode=cancer&mode_population=continents&population=900&sex=0&cancer=29&type=0&statistic=0&prevalence=0&color_palette=defaultSeptember 11–2018

2 

International Agency for Research on Cancer: Estimated number of deaths, both sexes, worldwide (top 10 cancer sites) in 2012. 2012.simplehttp://gco.iarc.fr/today/online-analysis-multi-bars?mode=cancer&mode_population=continents&population=900&sex=0&cancer=29&type=1&statistic=0&prevalence=0&color_palette=defaultSeptember 11–2018

3 

Wagner AD, Unverzagt S, Grothe W, Kleber G, Grothey A, Haerting J and Fleig WE: Chemotherapy for advanced gastric cancer. Cochrane Database Syst Rev. 3:CD0040642010.

4 

Hanahan D and Weinberg RA: Hallmarks of cancer: The next generation. Cell. 144:646–674. 2011. View Article : Google Scholar : PubMed/NCBI

5 

Szkandera J, Stotz M, Absenger G, Stojakovic T, Samonigg H, Kornprat P, Schaberl-Moser R, Alzoughbi W, Lackner C, Ress AL, et al: Validation of C-reactive protein levels as a prognostic indicator for survival in a large cohort of pancreatic cancer patients. Br J Cancer. 110:183–188. 2014. View Article : Google Scholar : PubMed/NCBI

6 

McMillan DC, Forrest LM, O'Gorman P, Angerson WJ and McArdle CS: Performance status of male and female advanced cancer patients is independently predicted by mid-upper arm circumference measurement. Nutr Cancer. 42:191–193. 2002. View Article : Google Scholar : PubMed/NCBI

7 

Wei Y, Jiang YZ and Qian WH: Prognostic role of NLR in urinary cancers: A meta-analysis. PLoS One. 9:e920792014. View Article : Google Scholar : PubMed/NCBI

8 

Imrie CW: Host systemic inflammatory response influences outcome in pancreatic cancer. Pancreatology. 15:327–330. 2015. View Article : Google Scholar : PubMed/NCBI

9 

Charlson ME, Pompei P, Ales KL and MacKenzie CR: A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis. 40:373–383. 1987. View Article : Google Scholar : PubMed/NCBI

10 

Sano T and Aiko T: New Japanese classifications and treatment guidelines for gastric cancer: Revision concepts and major revised points. Gastric Cancer. 14:97–100. 2011. View Article : Google Scholar : PubMed/NCBI

11 

Koizumi W, Narahara H, Hara T, Takagane A, Akiya T, Takagi M, Miyashita K, Nishizaki T, Kobayashi O, Takiyama W, et al: S-1 plus cisplatin versus S-1 alone for first-line treatment of advanced gastric cancer (SPIRITS trial): A phase III trial. Lancet Oncol. 9:215–221. 2008. View Article : Google Scholar : PubMed/NCBI

12 

Aapro MS, Köhne CH, Cohen HJ and Extermann M: Never too old? Age should not be a barrier to enrollment in cancer clinical trials. Oncologist. 10:198–204. 2005. View Article : Google Scholar : PubMed/NCBI

13 

Hutchins LF, Unger JM, Crowley JJ, Coltman CA Jr and Albain KS: Underrepresentation of patients 65 years of age or older in cancer-treatment trials. N Engl J Med. 341:2061–2067. 1999. View Article : Google Scholar : PubMed/NCBI

14 

Monfardini S, Sorio R, Boes GH, Kaye S and Serraino D: Entry and evaluation of elderly patients in European Organization for Research and Treatment of Cancer (EORTC) new-drug-development studies. Cancer. 76:333–338. 1995. View Article : Google Scholar : PubMed/NCBI

15 

Giovanazzi-Bannon S, Rademaker A, Lai G and Benson AB III: Treatment tolerance of elderly cancer patients entered onto phase II clinical trials: An Illinois Cancer Center study. J Clin Oncol. 12:2447–2452. 1994. View Article : Google Scholar : PubMed/NCBI

16 

Ajani JA, Rodriguez W, Bodoky G, Moiseyenko V, Lichinitser M, Gorbunova V, Vynnychenko I, Garin A, Lang I and Falcon S: Multicenter phase III comparison of cisplatin/S-1 with cisplatin/infusional fluorouracil in advanced gastric or gastroesophageal adenocarcinoma study: The FLAGS trial. J Clin Oncol. 28:1547–1553. 2010. View Article : Google Scholar : PubMed/NCBI

17 

Koizumi W, Akiya T, Sato A, Sakuyama T, Sasaki E, Tomidokoro T, Hamada T, Fujimori M, Kikuchi Y, Shimada K, et al: Phase II study of S-1 as first-line treatment for elderly patients over 75 years of age with advanced gastric cancer: The Tokyo Cooperative Oncology Group study. Cancer Chemother Pharmacol. 65:1093–1099. 2010. View Article : Google Scholar : PubMed/NCBI

18 

Imamura H, Kishimoto T, Takiuchi H, Kimura Y, Morimoto T, Imano M, Iijima S, Yamashita K, Maruyama K, Otsuji T, et al: Phase II study of S-1 monotherapy in patients over 75 years of age with advanced gastric cancer (OGSG0404). J Chemother. 26:57–61. 2014. View Article : Google Scholar : PubMed/NCBI

19 

Zhang X, Zhang W and Feng LJ: Prognostic significance of neutrophil lymphocyte ratio in patients with gastric cancer: A meta-analysis. PLoS One. 9:e1119062014. View Article : Google Scholar : PubMed/NCBI

20 

Hwang JE, Kim HN, Kim DE, Choi HJ, Jung SH, Shim HJ, Bae WK, Hwang EC, Cho SH and Chung IJ: Prognostic significance of a systemic inflammatory response in patients receiving first-line palliative chemotherapy for recurred or metastatic gastric cancer. BMC Cancer. 11:4892011. View Article : Google Scholar : PubMed/NCBI

21 

Namikawa T, Munekage E, Munekage M, Maeda H, Yatabe T, Kitagawa H, Kobayashi M and Hanazaki K: Evaluation of systemic inflammatory response biomarkers in patients receiving chemotherapy for unresectable and recurrent advanced gastric cancer. Oncology. 90:321–326. 2016. View Article : Google Scholar : PubMed/NCBI

22 

Arai T, Esaki Y, Inoshita N, Sawabe M, Kasahara I, Kuroiwa K, Honma N and Takubo K: Pathologic characteristics of gastric cancer in the elderly: A retrospective study of 994 surgical patients. Gastric Cancer. 7:154–159. 2004. View Article : Google Scholar : PubMed/NCBI

23 

Wang JY, Hsieh JS, Huang CJ, Huang YS and Huang TJ: Clinicopathologic study of advanced gastric cancer without serosal invasion in young and old patients. J Surg Oncol. 63:36–40. 1996. View Article : Google Scholar : PubMed/NCBI

24 

Hashimoto H, Esaki Y, Kino K, Takahashi T and Takubo K: Gastric cancer in the elderly - relationship between macroscopic appearance and histological type in autopsy specimens. Nippon Ronen Igakkai Zasshi. 34:499–505. 1997.(In Japanese). View Article : Google Scholar : PubMed/NCBI

25 

Mizoe A, Tanaka K, Ura K, Izawa K, Tsunoda T and Kanematsu T: Long term prognosis according to histological type of gastric cancer. J Jpn Pract Surg Soc. 54:1732–1737. 1993.(In Japanese).

26 

Forrest LM, McMillan DC, McArdle CS, Angerson WJ and Dunlop DJ: Evaluation of cumulative prognostic scores based on the systemic inflammatory response in patients with inoperable non-small-cell lung cancer. Br J Cancer. 89:1028–1030. 2003. View Article : Google Scholar : PubMed/NCBI

27 

Osugi J, Muto S, Matsumura Y, Higuchi M, Suzuki H and Gotoh M: Prognostic impact of the high-sensitivity modified Glasgow prognostic score in patients with resectable non-small cell lung cancer. J Cancer Res Ther. 12:945–951. 2016. View Article : Google Scholar : PubMed/NCBI

28 

Decoster L, Van Puyvelde K, Mohile S, Wedding U, Basso U, Colloca G, Rostoft S, Overcash J, Wildiers H, Steer C, et al: Screening tools for multidimensional health problems warranting a geriatric assessment in older cancer patients: An update on SIOG recommendations. Ann Oncol. 26:288–300. 2015. View Article : Google Scholar : PubMed/NCBI

Related Articles

Journal Cover

January 2019
Volume 10 Issue 1

Print ISSN: 2049-9450
Online ISSN:2049-9469

Sign up for eToc alerts

Recommend to Library

Copy and paste a formatted citation
APA
Sugimoto, A., Nishida, T., Osugi, N., Takahashi, K., Mukai, K., Nakamatsu, D. ... Inada, M. (2019). Prediction of survival benefit when deciding between chemotherapy and best supportive therapy in elderly patients with advanced gastric cancer: A retrospective cohort study. Molecular and Clinical Oncology, 10, 83-91. https://doi.org/10.3892/mco.2018.1772
MLA
Sugimoto, A., Nishida, T., Osugi, N., Takahashi, K., Mukai, K., Nakamatsu, D., Matsubara, T., Hayashi, S., Yamamoto, M., Nakajima, S., Fukui, K., Inada, M."Prediction of survival benefit when deciding between chemotherapy and best supportive therapy in elderly patients with advanced gastric cancer: A retrospective cohort study". Molecular and Clinical Oncology 10.1 (2019): 83-91.
Chicago
Sugimoto, A., Nishida, T., Osugi, N., Takahashi, K., Mukai, K., Nakamatsu, D., Matsubara, T., Hayashi, S., Yamamoto, M., Nakajima, S., Fukui, K., Inada, M."Prediction of survival benefit when deciding between chemotherapy and best supportive therapy in elderly patients with advanced gastric cancer: A retrospective cohort study". Molecular and Clinical Oncology 10, no. 1 (2019): 83-91. https://doi.org/10.3892/mco.2018.1772