You are here

Impact of age on the feasibility and efficacy of neoadjuvant chemotherapy in patients with locally advanced oesophagogastric cancer

European Journal of Cancer, Volume 51, Issue 14, September 2015, Pages 1918 - 1926



Neoadjuvant chemotherapy (neoCTx) improves the prognosis of patients with localised oesophagogastric adenocarcinoma (EGC), but its value is unknown in elderly patients.

Patients and methods

Patients who received neoCTx followed by surgery for EGC between 2000 and 2012 were analysed. The aim of this study was to compare the feasibility and outcome between patients aged ⩾70 (cohort I) and their younger counterparts (cohort II).


Data were available for 460 patients among which 174 (38%) were ⩾70 years. Older age was associated with an increased rate of comorbidities (66% versus 42%, p < 0,001). As compared to the younger, elderly patients were more likely to receive doublet instead of triplet neoCTx (65% versus 37%, p < 0.001) and oxaliplatin-instead of cisplatin-based regimens (60% versus 32%, p < 0.001). No significant difference was observed in the rate of ⩾grade 3 toxicities for cohort I and II (48% versus 41%) and postoperative morbidity was also not different (24% versus 28%). 90 day mortality for cohort I and II was 6.5% and 3.9%.

After a median follow-up of 38 months, median disease-free survival (DFS) was 29.4 months in cohort I and 33.8 months in cohort II, with a 5-years DFS of 37% and 40%, respectively. Median overall survival (OS) was not reached in cohort I and was 58.4 months in cohort II, with a 5-year OS of 51% and 50% for cohort I and II, respectively.


Despite slightly more adverse events and dose reductions, neoCTx is feasible in elderly patients with EGC. Elderly patients achieve comparable survival outcomes compared with their younger counterparts.

Keywords: Elderly patients, Toxicity, Oesophagogastric adenocarcinoma, Neoadjuvant chemotherapy.

1. Introduction

Oesophagogastric cancer (EGC) represents a challenging global health problem, with approximately 1 million cases per year occurring around the world. The incidence of EGC increases continuously with rising age. Currently, the median age at diagnosis is 70 years [1] .

As elderly patients tend to be under-represented in clinical trials, knowledge about treatment methods is mainly based on experience with patients younger than 70 years, and adequate data to allow robust conclusions regarding the tolerability and efficacy of specific regimens are lacking. In addition, particular concerns have emerged regarding the use of multimodal treatment strategies including major surgical procedures, such as oesophagectomy in elderly patients, leading to the exclusion of patients with an age ⩾65 or ⩾70 [2] .

There is a worldwide debate about the most appropriate definition of ‘elderly’ patients and the definition varies from 65 years or older to 70 years or older, however many experts favour a more functional definition of elderly patients, which is based on the patient’s functional health status or co-morbidities that may interfere with treatment decision making. In our analysis the age cut-off of ⩾70 years was chosen as it defines ‘elderly’ throughout a large portion of the epidemiological literature.

Based on the results of two randomised phase III trials, preoperative chemotherapy is the standard of care for locally advanced EGC in large parts of the western world, leading to an estimated improvement in the 5-year survival rate of 13% compared to surgery alone [3] and [4]. Of note, in the MAGIC trial [3] the median age was 62 years (range 23–85) and only 20% were ⩾70 years. Clinical trials in general only include the fittest older adults, yielding data that cannot be generalised to a heterogeneous older cohort. Therefore, there are valid concerns regarding the use of neoadjuvant chemotherapy in elderly patients. On the other hand such age-based treatment biases may be detrimental to many fit elderly patients if they lead to appropriate cancer therapies being denied [5] . As data on feasibility, toxicities and efficacy of neoadjuvant chemotherapy in older patients are lacking, this retrospective analysis addressed the following questions:

  • 1. Is neoadjuvant chemotherapy feasible in older patients and do they experience a different pattern or severity of adverse events?
  • 2. Do elderly patients have comparable rates of postoperative morbidity and mortality after neoadjuvant chemotherapy and surgical resection of their primary tumour?
  • 3. Do patients older than 70 years derive the same benefit from neoadjuvant chemotherapy as assessed by means of pathological response and DFS?

2. Patients and methods

Patients who underwent neoadjuvant chemotherapy followed by surgery with curative intent for EGC between 2000 and 2012 from three institutions of the Arbeitsgemeinschaft Internistische Onkologie (AIO) steering group were analysed. On total, 192 of the 460 analysed patients had participated in three clinical trials [6], [7], and [8]. All patients had histologically proven localised (cT3 or N+) EGC with no evidence of distant metastases. Medical records included surgical and pathological records, imaging records, chemotherapy protocols showing dose reductions and delays, patient’s charts where toxicity was documented and follow-up reports informing about date of progression, last follow-up and death. Toxicity was graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAEs) in its version valid at the time when the respective patients were treated. The relative safety and efficacy of neoadjuvant chemotherapy was compared in two cohorts: those aged ⩾70 (cohort I, n = 174) were compared with their younger counterparts (cohort II, n = 286) in terms of co-morbidities, treatment delivery (type, dose reductions, early discontinuation of therapy), chemotherapy-associated toxicity (incidence of major toxic events ⩾grade 3), postoperative morbidity, 90 day postoperative mortality and survival outcomes (overall survival [OS], disease-free survival [DFS]). In all four institutions, neoadjuvant chemotherapy was considered the standard of care in potentially resectable tumours with clinical stages cT2/3, N+, M0.

2.1. Tumour staging

Comparable staging procedures were used among the participating centres. Clinical staging consisted of endoscopy including endoluminal ultrasound and computed tomography scans of the chest, abdomen and pelvis to assess the clinical T- and N-stages, and were repeated before surgery to exclude disease progression. All patients underwent definitive resection and, therefore, had final tumour histology available for comparison (ypTNM). The pathologic stage was defined using the seventh edition of the American Joint Committee on Cancer TNM staging system [9] , including R (residual tumour) category and grading. Downstaging was defined as a reduction in T stage or N stage of pathologic staging (ypTNM) compared with clinical staging (cTNM).

Histopathological evaluation was performed at the respective centre by standardised protocols including the pTNM categories, tumour localisation, subtype according to Lauren’s classification and resection margins, as demanded in the guidelines of the UICC [10] .

Tumour regression was assessed semi-quantitatively according to a previously published scoring system by Becker [11] . Pathologic tumour regression used a categorical scale between 1a (no residual tumour cells) and 3 (>50% residual tumour/tumour bed) [11] and [12].

2.2. Follow-up

Participating centres used comparable follow-up schedules. Patients who had ended treatment but had not experienced disease progression were observed every 3 months in the first two years, followed by 6 months intervals until disease progression. DFS was measured from the start of chemotherapy until disease progression or death of any cause; OS was calculated from the start of chemotherapy until death of any cause.

2.3. Data collection

All patient data including preoperative, operative, postoperative and follow-up information were collected prospectively and entered into a computerised database with prospectively defined database elements and monthly quality assurance review. Details of the specific chemotherapy regimens and dose reductions were not part of the database; therefore this information was collected in a retrospective fashion from the medical charts. Details of why individual patients did or did not undergo preoperative chemotherapy were not reliably recorded in the medical records and therefore were not included in the analysis.

2.4. Statistical analysis

Patients ⩾70 years and patients <70 years were compared regarding categorical and time-to-event data. Categorical data are presented as absolute and relative frequencies. Comparisons of patient groups were conducted using chi-squared tests or Fisher’s exact tests, depending on the expected cell counts under the null hypothesis. Distributions of overall and disease-free survival were estimated using the Kaplan–Meier method. Estimates for median survival and for the 3-year- and 5-year overall survival and disease-free survival probabilities are presented with 95% confidence intervals. Hazard rates were compared between the groups using the log-rank test. Cox proportional hazards models were fit to the data to derive hazard ratios with 95% confidence intervals and to estimate the hazard ratio between both age groups adjusted for possible confounders (presence of comorbidities, gender, site of tumour, number of drugs given as neoadjuvant chemotherapy). Median follow-up was estimated using the inverse Kaplan–Meier estimator for potential follow-up [13] . A two-sided level of significance of 5% was used for all tests. Statistical analysis was performed using R version 3.1.1 and SPSS version 22.

3. Results

From January 2000 through December 2012 460 patients with locally advanced EGC cancer were treated with neoadjuvant chemotherapy followed by resection. Using the cut-off of 70 years, 174 (38%) were assigned to the elderly (cohort I) and 286 (62%) to the non-elderly group (cohort II). The median age in cohort I and II was 74 (range 70–85) and 60 (range 33–69) years, respectively. Demographic data are summarised in Table 1 .

Table 1 Demographic and pre-therapeutic data.

  Cohort I (⩾70 years age) Cohort II (<70 years age) p-value
n = 174 %38 n = 286 %62
Gender         0.033
Male 134 77 244 85.3  
Female 40 23 42 14.7  
Age groups          
<70     286 62  
70–74 111 25      
75–79 57 12      
>80 6 1      
Comorbidities ⁎⁎⁎ 107 65.6 104 42.3 <0.001
Coronary artery disease 40 24.8 31 12.7  
Hypertension 84 51.5 79 32.1  
History of myocardial infarction 25 15.4 22 9  
Arrhythmia 20 12.3 9 3.7  
Diabetes 23 14.2 20 8.1  
COPD 15 9.2 8 3.3  
Prior stroke 4 2.5 7 2.8  
PAOD ⁎⁎ 5 3.1 4 1.6  
Renal insufficiency 4 2.5 5 2  
Site of tumour          
GEJ 1 105 64.4 202 74.8 0.028
Stomach 58 35.6 68 25.2  
Histological type          
Diffuse 20 18.2 34 21.5 0.630
Intestinal 77 70 110 69.6  
Mixed 13 11.8 14 8.9  
Tumour grading          
G1 4 2.6 4 1.8 0.824
G2 46 30.5 74 32.6  
G3 96 63.6 145 63.9  
G4 5 3.3 4 1.8  
Clinical stage          
uT1/2 1 0.6 6 2.1 0.367
uT3/4 173 99.4 280 97.9
N0 2 1.1 5 1.7 0.909
N+ 172 98.9 281 98.3

Chronic obstructive pulmonary disease.

⁎⁎ Peripheral arterial occlusive disease.

⁎⁎⁎ Data available for 410 patients.

1 Gastroesophageal junction (GEJ).

Older age was associated with an increased rate of comorbidities (65.6% versus 42.3%, p < 0.001). Subgroup analysis found that patients aged ⩾75 had an even higher rate of comorbidities (73.4%) compared to patients aged 70–74 (60.6%) and patients in cohort II. Distribution according to histopathological subtype of the tumour, grading and clinical stage was not different between the two groups.

3.1. Preoperative treatment

Of the 460 patients who started neoadjuvant chemotherapy, at least one dose attenuation to <80% was necessary in 26.6% and 20.2% in cohort I and II, respectively (p = 0.144) ( Table 1 supplements online only ).

Preoperative chemotherapy was not completed due to toxic effects in 17.8% of the elderly patients and 10.2% of the younger ones (p = 0.029). Rates of ⩾grade 3 adverse events by age group are listed in Table 2 .

Table 2 Adverse events (National Cancer Institute Common Terminology Criteria for Adverse Event (CTCAE) grade ⩾ 3) during neoadjuvant chemotherapy.

Major toxic events (National Cancer Institute Common Terminology Criteria for Adverse Event (CTCAE) ⩾ grade 3) Group I (⩾70) Group II (<70) p-value
n = 174 % n = 279 %
all ⁎⁎ 84 48.3 115 41.2 0.169
Haematologic 22 12.6 35 12.5 0.909
Neutropenia 21 12 34 12.1  
Thrombocytopenia 1 0.6 1 0.4  
Non-haematologic 62 35.6 80 28.6 0.147
Diarrhoea 13 7.5 15 5.3  
Renal toxicity 7 4 5 1.8  
Polyneuropathy 8 4.6 5 1.8  
Nausea/vomiting 3 1.7 12 4.3  
Allergic reaction 1 0.6 0 0  
Asthenia 2 1.2 7 2.5  
Ototoxicity 1 0.6 0 0  
Cardiotoxicity 3 1.7 3 1  
Hand-foot-syndrome 0 0 2 0.7  
Other 24 12.6 31 11  

Data available for 279 patients.

⁎⁎ One patient may have experienced more than one ⩾ grade 3 adverse event. This is why the sum of single events is more than the number given as ‘all’.

Overall, grade 3/4 toxicity occurred in 84 of 174 patients in cohort I (48.3%) and 115 of 279 patients (41.2%) in cohort II (p = 0.169). Patients aged ⩾75 had a higher incidence of grade ⩾3 toxicities (56.3%) compared to patients aged 70–74 (43.6%) and patients in cohort II.

The rates of haematological toxicity did not differ by age groups nor did the rate of non-haematological side-effects.

As compared to the younger, elderly patients were more likely to receive doublet chemotherapy compared to triple-agent regimens (65% versus 37%, p < 0.001) and oxaliplatin-containing regimens compared to cisplatin-based regimens (60% versus 32%; p < 0.001) ( Table 2 supplements online only ).

3.2. Surgery

All patients included in the analysis underwent resection of the primary tumour. Surgery consisted of total gastrectomy (n = 63), abdomino-thoracic oesophagectomy (Ivor-Lewis-procedure, RTA) (n = 211), subtotal gastrectomy (n = 20), total gastrectomy with transhiatal extension to the distal oesophagus (TH) (n = 93), transmediastinal (n = 15) and unknown (n = 58). The distribution of performed surgical procedure did not differ between both cohorts. However, when the subgroup of patients involving resection of the distal oesophagus (TH or RTA) was investigated (n = 304), RTA was more frequently performed in younger patients (85.8% versus 14.2%), whereas TH was more commonly performed in the elderly (57.9% versus 52.1%) (p < 0.001). When OS was analysed in this subgroup, no significant difference could be observed between the different surgical procedures in the younger (p = 0.136) and the elderly patients (p = 0.134).

Overall the R0 resection rate was 89.5%, with 86.4% and 91.3% in cohort I and II respectively (R-status was available for 438 patients). Positive lymph nodes out of removed lymph nodes during surgery in cohort I and II were 3/25.3 and 2.7/27 (p = 0.037).

The incidence of postoperative complications was similar in the two cohorts (24.1% in cohort I and 28.3% in cohort II; p = 0.382), as was the 90 day mortality rate (6.5% in cohort I and 3.9% in cohort II, p = 0.21). Pulmonary complications were the most common encountered event but there were no significant differences between the groups. Table 3 describes frequencies of postoperative complications for both groups.

Table 3 Postoperative complications.

Type of complication Operated patients
Cohort I (⩾70 years age) Cohort II (<70 years age) p-value
n = 174 % n = 286 %
All ⁎⁎ 42 24.1 81 28.3 0.382
Anastomotic leakage 5 2.9 17 5.9 0.204
Pulmonary (pneumonia, lung failure, pleural effusion) 15 8.6 27 9.4 0.896
Cardiovascular (myocardial infarction, arrhythmia) 5 2.9 1 0.4 0.058
Bleeding 3 1.7 6 2 0.949
Enterothorax 2 1.1 1 0.4 0.662
Wound infection 3 1.7 14 4.9 0.135
Fever 4 2.2 4 1.3 0.727
Other 3 1.7 11 3.8 0.314

⁎⁎ One patient may have experienced more than one ⩾ grade 3 adverse event. This is why the sum of single events is more than the number given as ‘all’.

We compared rates of major postoperative adverse events and overall survival between elderly patients treated with doublet and triplet therapy. Neither rates of postoperative complications (doublet: 22%, triplet: 28%, p = 0.464) nor overall survival (hazard ratio (HR) = 0.66, 95% confidence interval (CI): 0.37–1.16, p = 0.143) were significantly associated with the number of drugs delivered.

3.3. Response to neoadjuvant chemotherapy

Of 430 evaluable patients 125 patients in cohort I (74.9%) and 181 patients in cohort II (68.8%) benefitted from a downstaging effect ( Table 3 supplements online only ).

The majority of downstaged patients in both cohorts had evidence of pathologic response to neoadjuvant chemotherapy ( Table 4 supplements online only ). Complete histopathological response was seen in 13 patients in cohort I (7.8%) and 31 patients (11.7%) in cohort II, with a higher proportion of patients achieving a pCR with triple therapy.

The likelihood of histopathological response and tumour downstaging was not significantly associated with age (p = 0.396).

Patients in cohort I with downstaging had a 3-year survival rate of 66.4% (57.3–76.8%) compared with 30.1% (17.0–53.3%) in non-downstaged elderly patients. For patients in cohort II the 3-year survival rate in downstaged patients was slightly higher with 71.7% (64.7–79.6%) compared to 50.4% (39.8–63.7%) in the non-downstaged population of cohort II.

Fig. 1 represents the estimated survival curves of the patients initially staged cT3/4N+ and their outcome according to downstaging or non-downstaging ( Fig. 1 a for cohort I; Fig. 1 b for cohort II). Fig. 1a and b supplements online only represents a more detailed survival analysis according to age group of the patients initially staged as cT3/4N+ who proceeded to chemotherapy and were downstaged or not downstaged.


Fig. 1 (a and b) Comparison of surgically treated patients by downstaging effect according to age group (1a cohort I; 1b cohort II).

3.4. Survival

The median follow-up for all patients was 38.0 months (minimum 2.5 months; maximum 60.0 months) from the date of chemotherapy start, with a median follow-up of 32.8 months in cohort I and 38.8 months in cohort II, respectively. During follow-up 172 deaths were observed. Median DFS was 29.4 months in cohort I and 33.8 months in cohort II, with a 5-years DFS of 36.8% (95% CI, 29.0–46.8%) and 40.0% (95% CI, 33.6%–47.6%) for cohort I and II, respectively (p = 0.302). Median OS was not reached in cohort I and was 58.4 months for cohort II and 5-years OS was 51.0% (95% CI, 42.3–61.4%) and 50.0%, (95% CI, 42.9–58.2%), for cohort I and II respectively (p = 0.375) ( Fig. 2 ). Of note, 5-years DFS and 5 years OS of patients younger than 70, 70–74 and 75 and older were comparable. On multivariable Cox regression, age was not significantly associated with overall survival after adjustment for the presence of co-morbidities, gender, site of tumour and the number of chemotherapy agents applied (HR for age = 0.87, 95% CI: 0.59–1.27. p = 0.462). After adjustment for site of tumour, younger age was associated with a slightly better overall- (HR: 0.82; 95% CI 0.60–1.13; p = 0.23) and disease-free survival (HR: 0.83; 95% CI 0.63–1.08; p = 0.16).


Fig. 2 Kaplan–Meier plots of overall survival (2a) and disease-free survival (2b) by age group.

4. Discussion

Using source data of 460 patients with locally advanced EGC, we identified 174 patients aged 70 years or older, representing 38% of the entire cohort. This does not reflect the proportion of patients with EGC that are aged ⩾70 years and may rather reflect physician’s bias, considering neoadjuvant chemotherapy in older patients with locally advanced disease less frequently. The MAGIC trial stratified patients by age including around 20% of patients >70 years. The authors report that there was no clear evidence of heterogeneity of treatment effect depending on age (P = 0.43), unfortunately they do not specify on the feasibility [3] . The FFCD study included patients until the age of 75, but reports on feasibility and efficacy in the cohort of elderly patients are lacking [4] .

Apart from a significantly increased rate of co-morbidities in elderly patients (66% versus 42%), there were only minor differences between patients ⩾70 years and those <70 years, including a male predominance and relatively more tumours located at the gastroesophageal junction in the younger patient cohort. This, and the fact that surgeons were more inclined to choose a less invasive procedure in older patients explains the higher rate of Ivor-Lewis procedures in the younger patients compared to the elderly in the subgroup of patients who underwent procedures involving oesophageal resections.

The present study provides relevant information regarding the safety and feasibility of preoperative chemotherapy in elderly patients. In the elderly population, 73% of patients completed preoperative treatment without dose adjustments; however, toxic effects were more frequent, with almost half of the patients developing ⩾3 grade toxicities leading to a significantly higher rate of treatment interruptions (18% versus 10%). Of note, no significant differences in severe (grade 3–4) haematological and non-haematological toxicities were noticed between elderly and non-elderly patients, although this may be, at least in part, attributed to the lower intensity of chemotherapy. Additionally, type II errors cannot be excluded for some of the statistical tests performed. Bias may be evident that patients less than 70 years old were much more likely to receive three-drug combinations of chemotherapy, while older patients generally received two-drug combinations. Elderly patients are more likely to receive oxaliplatin-based regimens, which are, compared to cisplatin-based therapy, less toxic but similarly active in EGC, and might be better suited for this population [14] .

In terms of feasibility we did not register an increased rate of postoperative morbidity and mortality in elderly patients despite an increased rate of comorbidities in this cohort. The overall postoperative morbidity rate of 24% in elderly patients compared well with those reported in the MAGIC and the FFCD trial, where a 45% and 28% morbidity rate was shown, and did not differ significantly between patients who received neoadjuvant doublet- or triplet chemotherapy regimens. These results clearly point towards that elderly patients receiving neoadjuvant therapy are not burdened by a significantly higher risk of developing major or fatal postoperative complications as compared to their younger counterparts [15] and [16].

Despite the lower intensity of chemotherapy with less triplet agent regimens administered, and more adverse advents and dose reductions due to toxicities in the elderly patient population, the likelihood of response and tumour downstaging was not significantly associated with age (p = 0.396). Elderly patients had comparable pathological complete remission rates compared to the younger ones (8% versus 11%). As demonstrated in other studies, histopathological response to preoperative treatment is predictive for improved DFS [17] and [18] and correlates with long-term survival [19] . Therefore, elderly patients do not seem to have a major disadvantage from treatment with less intense regimens.

Efficacy of chemotherapy did not differ when patients ⩾70 years were compared to younger ones. The median disease free-survival of elderly patients was not significantly different from that of younger patients (29 versus 34 months) as was the median overall survival (not reached versus 58 months). Similar results have been recently reported for neoadjuvant radiochemotherapy in oesophageal cancer [20] . Indeed, multivariate analysis confirmed that age ⩾70 years was not a significant prognostic factor for survival. The absence of a negative influence of age on chemotherapy efficacy has been shown previously in a selected population of elderly patients with palliative chemotherapy in EGC [21] . One limitation of the study is that we did not perform a cancer specific survival analysis because reliable information about the cause of death could not be comprehensively retrieved in this retrospective analysis. Furthermore, due to the retrospective design of the study, some desirable information was not available, including data on performance status and quality of life. Inability to adjust for these variables in multivariable analysis may have caused residual confounding. Furthermore, a comprehensive geriatric assessment (CGA) [22] to identify functional limitations of the elderly as well as patients with increased risk of adverse outcomes, was not applied. Prospective trials incorporating measures of factors beyond physiologic age (eg. comorbid conditions, patient functional status, or dose intensity, as collected by the Cancer-Specific Geriatric Assessment) [23] may guide clinicians in optimal treatment selection for older patients, limiting the potential lack of benefit or harm to vulnerable or frail older patients. Another limitation of the analysis is that only patients who underwent neoadjuvant chemotherapy followed by surgery were included in the analysis as opposed to an intention-to-treat analysis. Therefore we do not know the proportion of patients who started chemotherapy and subsequently were not taken to surgery because of complications or disease progression or patients who directly went into surgery. Other trials reported on a dropout rate of up to 25% prior to surgery in a selected patient population with a median age of 70 years [24] . Compared to the MAGIC and FFCD trial where patients with a median age of 62 and 63 years who were randomised to perioperative chemotherapy subsequently underwent surgery in 91.6% and 96.5%, respectively, the dropout rate seems to be relevantly increased in our population of elderly patients. Usually elderly patients undergoing major surgical procedures are a selected population, which has undergone a careful evaluation of physiological functions. A decreased likelihood of having surgery has been reported with increasing age in several studies [25], [26], and [27], which may be a reflection of the presence of toxicity and comorbidity, also influencing patient’s choice. It is therefore noticeable, that the elderly patients included in this analysis may represent a group of relatively fit elderly patients and, therefore, it may underestimate what might have been actually seen in a prospective randomised trial. Furthermore, the survival curves for elderly and younger patients demonstrate that age does not appear to be a prognostic factor for survival. Due to the above mentioned bias this analysis does not answer the question if similar survival curves could have been achieved for elderly patients undergoing surgery alone; this has to be evaluated prospectively.

In conclusion, this analysis is, to our knowledge, the largest report on neoadjuvant chemotherapy in elderly patients with EGC. The clinical relevance of our study is that elderly patients did not experience an excess of severe toxicity and morbidity due to neoadjuvant chemotherapy followed by surgery. However, we acknowledge that the overall strategy in elderly patients is only feasible under consideration of some specificities such as an appropriate chemotherapy regimen, lower dose intensity, intensive surveillance of toxicity occurrence and their management and an a priori careful patient selection. Further studies are needed to investigate the reproducibility of the present results in independent patient cohorts or in prospective trials.

All aforementioned procedures were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and later versions.

Conflict of interest statement

None declared.

Appendix A. Supplementary data


Download file

Supplementary Fig. 1 Survival analysis of the patients initially staged cT3/4N+ according to downstaging or non-downstaging after neoadjuvant chemotherapy and resection in cohort I (1a) and cohort II (1b).

Download file

Supplementary Table 1 Dose reductions and discontinuation of neoadjuvant chemotherapy.

Download file

Supplementary Table 2 Neoadjuvant chemotherapy regimen.

Download file

Supplementary Table 3 Tumour downstaging after neoadjuvant chemotherapy in patients undergoing surgery for EGC.

Download file

Supplementary Table 4 Histopathologic response to neoadjuvant chemotherapy.


  • [1] SEER. Cancer Statistics 18 2007–2011.
  • [2] L.F. Hutchins, J.M. Unger, J.J. Crowley, C.A. Coltman Jr., K.S. Albain. Underrepresentation of patients 65 years of age or older in cancer-treatment trials. N Engl J Med. 1999;341(27):2061-2067 Crossref
  • [3] D. Cunningham, W.H. Allum, S.P. Stenning, J.N. Thompson, C.J. Van de Velde, M. Nicolson, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med. 2006;355(1):11-20 Crossref
  • [4] M. Ychou, V. Boige, J.P. Pignon, T. Conroy, O. Bouche, G. Lebreton, et al. Perioperative chemotherapy compared with surgery alone for resectable gastroesophageal adenocarcinoma: an FNCLCC and FFCD multicenter phase III trial. J Clin Oncol. 2011;29(13):1715-1721 Crossref
  • [5] J.S. Goodwin, J.M. Samet, W.C. Hunt. Determinants of survival in older cancer patients. J Natl Cancer Inst. 1996;88(15):1031-1038 Crossref
  • [6] S. Lorenzen, M. Hentrich, C. Haberl, V. Heinemann, T. Schuster, T. Seroneit, et al. Split-dose docetaxel, cisplatin and leucovorin/fluorouracil as first-line therapy in advanced gastric cancer and adenocarcinoma of the gastroesophageal junction: results of a phase II trial. Ann Oncol. 2007;18(10):1673-1679 Crossref
  • [7] P.C. Thuss-Patience, R.D. Hofheinz, D. Arnold, A. Florschutz, S. Daum, A. Kretzschmar, et al. Perioperative chemotherapy with docetaxel, cisplatin and capecitabine (DCX) in gastro-oesophageal adenocarcinoma: a phase II study of the Arbeitsgemeinschaft Internistische Onkologie (AIO){dagger}. Ann Oncol. 2012;23(11):2827-2834 Crossref
  • [8] S.E. Al-Batran, C. Pauligk, N. Homann, J.T. Hartmann, M. Moehler, S. Probst, et al. The feasibility of triple-drug chemotherapy combination in older adult patients with oesophagogastric cancer: a randomised trial of the Arbeitsgemeinschaft Internistische Onkologie (FLOT65+). Eur J Cancer. 2013;49(4):835-842 Crossref
  • [9] T.W. Rice, E.H. Blackstone, V.W. Rusch. 7th edition of the AJCC Cancer Staging Manual: esophagus and esophagogastric junction. Ann Surg Oncol. 2010;17(7):1721-1724 Crossref
  • [10] F.L. Greene, et al. AJCC Cancer Staging Manual. 6th ed. (Springer, New York, 2002)
  • [11] K. Becker, J.D. Mueller, C. Schulmacher, K. Ott, U. Fink, R. Busch, et al. Histomorphology and grading of regression in gastric carcinoma treated with neoadjuvant chemotherapy. Cancer. 2003;98(7):1521-1530 Crossref
  • [12] K. Becker, R. Langer, D. Reim, A. Novotny, C. Meyer zum Buschenfelde, J. Engel, et al. Significance of histopathological tumor regression after neoadjuvant chemotherapy in gastric adenocarcinomas: a summary of 480 cases. Ann Surg. 2011;253(5):934-939 Crossref
  • [13] M. Schemper, T.L. Smith. A note on quantifying follow-up in studies of failure time. Control Clin Trials. 1996;17(4):343-346 Crossref
  • [14] S.E. Al-Batran, J.T. Hartmann, S. Probst, H. Schmalenberg, S. Hollerbach, R. Hofheinz, et al. Phase III trial in metastatic gastroesophageal adenocarcinoma with fluorouracil, leucovorin plus either oxaliplatin or cisplatin: a study of the Arbeitsgemeinschaft Internistische Onkologie. J Clin Oncol. 2008;26(9):1435-1442 Crossref
  • [15] A. Ruol, G. Portale, C. Castoro, S. Merigliano, M. Cagol, F. Cavallin, et al. Effects of neoadjuvant therapy on perioperative morbidity in elderly patients undergoing esophagectomy for esophageal cancer. Ann Surg Oncol. 2007;14(11):3243-3250 Crossref
  • [16] F.C. Lin, A.E. Durkin, M.K. Ferguson. Induction therapy does not increase surgical morbidity after esophagectomy for cancer. Ann Thorac Surg. 2004;78(5):1783-1789 Crossref
  • [17] D.P. Kelsen, K.A. Winter, L.L. Gunderson, J. Mortimer, N.C. Estes, D.G. Haller, et al. Long-term results of RTOG trial 8911 (USA Intergroup 113): a random assignment trial comparison of chemotherapy followed by surgery compared with surgery alone for esophageal cancer. J Clin Oncol. 2007;25(24):3719-3725 Crossref
  • [18] J.C. Mansour, L. Tang, M. Shah, D. Bentrem, D.S. Klimstra, M. Gonen, et al. Does graded histologic response after neoadjuvant chemotherapy predict survival for completely resected gastric cancer?. Ann Surg Oncol. 2007;14(12):3412-3418 Crossref
  • [19] S. Lorenzen, P. Thuss-Patience, S.E. Al-Batran, F. Lordick, B. Haller, T. Schuster, et al. Impact of pathologic complete response on disease-free survival in patients with esophagogastric adenocarcinoma receiving preoperative docetaxel-based chemotherapy. Ann Oncol. 2013;24(8):2068-2073 Crossref
  • [20] D.C. Rice, A.M. Correa, A.A. Vaporciyan, N. Sodhi, W.R. Smythe, S.G. Swisher, et al. Preoperative chemoradiotherapy prior to esophagectomy in elderly patients is not associated with increased morbidity. Ann Thorac Surg. 2005;79(2):391-397 Crossref
  • [21] M. Trumper, P.J. Ross, D. Cunningham, A.R. Norman, R. Hawkins, M. Seymour, et al. Efficacy and tolerability of chemotherapy in elderly patients with advanced oesophago-gastric cancer: a pooled analysis of three clinical trials. Eur J Cancer. 2006;42(7):827-834 Crossref
  • [22] P. Caillet, F. Canoui-Poitrine, J. Vouriot, M. Berle, N. Reinald, S. Krypciak, et al. Comprehensive geriatric assessment in the decision-making process in elderly patients with cancer: ELCAPA study. J Clin Oncol. 2011;29(27):3636-3642 Crossref
  • [23] A. Hurria, K. Togawa, S.G. Mohile, C. Owusu, H.D. Klepin, C.P. Gross, et al. Predicting chemotherapy toxicity in older adults with cancer: a prospective multicenter study. J Clin Oncol. 2011;29(25):3457-3465 Crossref
  • [24] S. Lorenzen, C. Pauligk, N. Homann, H. Schmalenberg, E. Jager, S.E. Al-Batran. Feasibility of perioperative chemotherapy with infusional 5-FU, leucovorin, and oxaliplatin with (FLOT) or without (FLO) docetaxel in elderly patients with locally advanced esophagogastric cancer. Br J Cancer. 2013;108(3):519-526 Crossref
  • [25] M.S. Sabel, J.L. Smith, H.R. Nava, K. Mollen, H.O. Douglass, J.F. Gibbs. Esophageal resection for carcinoma in patients older than 70 years. Ann Surg Oncol. 2002;9(2):210-214
  • [26] D.P. Cronin-Fenton, L. Sharp, A.E. Carsin, H. Comber. Patterns of care and effects on mortality for cancers of the oesophagus and gastric cardia: a population-based study. Eur J Cancer. 2007;43(3):565-575 Crossref
  • [27] L.B. Koppert, V.E. Lemmens, J.W. Coebergh, E.W. Steyerberg, B.P. Wijnhoven, H.W. Tilanus, et al. Impact of age and co-morbidity on surgical resection rate and survival in patients with oesophageal and gastric cancer. Br J Surg. 2012;99(12):1693-1700 Crossref


a 3rd Department of Internal Medicine (Hematology/Medical Oncology), Klinikum rechts der Isar, Technische Universität München, Munich, Germany

b Krankenhaus Nordwest, UCT University Cancer Center, Frankfurt am Main, Germany

c Department of Surgery, Klinikum rechts der Isar der Technischen Universitaet Muenchen, Muenchen, Germany

d University Cancer Center Leipzig (UCCL), University Medicine Leipzig, University of Leipzig, Leipzig, Germany

e Department of Haematology, Oncology and Tumorimmunology, Campus Virchow-Klinikum, Charite-University Medicine Berlin, Berlin, Germany

f Institute for Medical Statistics and Epidemiology, Technische Universität München, Munich, Germany

g Department of General, Visceral, Transplantation, Vascular and Thoracic Surgery, Hospital of the University of München, Munich, Germany

Corresponding author at: 3rd Department of Internal Medicine (Hematology/Medical Oncology), Technical University of Munich, Ismaninger Straße 22, D-81675 Munich, Germany. Tel.: +49 89 4140 9848; fax: +49 89 4140 4822.

1 Contributed equally.