With the extension of life expectancy, more ES patients would present in their ages over 60 years old.(18) A Chinese study indicated an increased incidence and mortality was observed among the patients older than 70 years comparing to other age stratifications,(19) leading to a sharp increase health care costs in both the developing(20) and developed(21) countries. However, the guidelines that specifically address the management of the older EC patient are rare, only a few studies referring the topics of evaluation and treatment for the elderly.
In the present study, based on the cohorts after PSM, we found that both chemotherapy and radiotherapy could present a positive impact on CSD and OS of the elderly patients. Then, the univariate and multivariate competing regression model were performed to identify lower grade, advanced T-stage and positive lymph node status were significantly correlated to the higher probabilities of CSD. Besides, the tumor size and distant metastasis were also considered as the key factors to predict the prognosis of EC. Kamel et.al. pointed out the size of malignancy (HR = 1.005) was a significantly independent predictor to the CSD for the T1N0M0 patients based on the SEER database.(22) Malnutritional status were correlated with poor survival,(23) while the tumor size of esophagus was significantly correlated to nutritional status measured by prognostic nutritional index (p = 0.016).(24) Furthermore, larger tumor size was a predictive factor to identify the EC patients who might have a higher rate of resection with the help of neoadjuvant therapy.(25) When it came to the tumor site, the different anatomical locations of EC was usually correlated to the variable treatment responses as well as survival outcome. The upper third was commonly associated with the poor prognosis, but tended to be more sensitivity to the chemo-radiotherapy.(26) Finally, after reviewing 838 patients with EC between 1982 and 1993, Quint indicated metastases were commonly diagnosed in lymph nodes, liver, lung, bone, adrenal, etc., determining the further management and predicting prognosis.(27)
By the variables mentioned above, a nomogram could be established to estimate the risk score for every patient, then the high- and low- risk (score) groups could be derived from whole-population cohort. With a series of analysis, both the chemotherapy and radiotherapy were showed to decrease the probabilities of CSD as well as improve the survival in the high-risk group, indicating elderly patients with the disease characterized as middle- & upper- third of tumor location, poorly- & un-differentiated of histological grade, >=35 mm of tumor size, advanced T-stage, positive lymph node and (or) metastasis tend to benefit from the relative aggressive treatment. However, in the low-risk group, no significant benefits for CSD were observed in the elderly population neither receiving chemotherapy nor radiotherapy. Furthermore, by performing the log-rank test, a significantly negative impact of radiotherapy was observed on the OS, suggesting providing aggressive therapies for the old characterized as low-third of tumor location, well- & moderate-differentiated of histological grade, < 35 mm of tumor size, early T-stage, negative lymph node and (or) non-metastasis was unfavorable. The above result was concordance with the current studies as follows.
As for the chemotherapy, the OE0-2 trial conducted by Medical Research Council of the U.K showed, receiving regimens of cisplatin and 5-FU before surgery (n = 86) could prolong the survival of patients (> 75y) with a HR of 0.7, compared to the ones receiving surgery alone (n = 79).(28) Besides, a randomized trial from Germany indicated that, compared to the double combinations of 5-fluorouracil, leucovorin and oxaliplatin (FLO) a triple combination of 5-fluorouracil, leucovorin, oxaliplatin and docetaxel (FLOT) could improve therapeutic response and progression-free survival in the patients aged from 65 to 70 years, which was associated with higher incidence of side effects for diarrhea (P = 0.006), alopecia (P < 0.001), neutropenia (P < 0.001), nausea (P = 0.029) and leukopenia (P < 0.001).(29) When it comes to the radiotherapy, the CROSS trail found that chemoradiotherapy followed surgery was more favorable for patients with a median age of 60y, compared to the surgery alone.(30) The median OS was 48.6 months versus 24.0 months (HR: 0·68, P = 0·003), respectively. However, a multi-center randomized phase III trial of FFCD 9901 containing 195 patients with a median age of approximately 60 years old to study the effect of neoadjuvant chemoradiotherapy on the ones with early stage EC, as a result, no significant positive impact on rate of R0 resection or survival prognosis was observed, but the mortality after surgery was increased.(31)
Generally speaking, it was widely accepted that both the chemotherapy and radiotherapy could exert a positive effect on the survival of elderly EC patients. For instance, after reviewing 21593 EC patients aged 70 years or older from the National Cancer Database, Gregory et.al pointed out any cancer-related treatment could play a positive role to prolong the survival of the old EC patient.(32) Similarly, Daniela et.al used SEER Database(2001–2009) to show that an improved 5-year survival could be observed in the elder patients(≥ 65y) receiving any medical or surgical therapy.(33) But, the possible defect in these studies was that the authors did not use competitive risk model to avoid the interference of non-cancer specific death on the survival analysis. Furthermore, with consideration of the balance between benefits and harms referring therapies in the old population, it was vital to identify the specific crowd tended to benefit or be suffered from the anti-tumor treatment, which was usually absent in most current studies.
There were some limitations in our studies. Firstly, it was a retrospective research based on the SEER only covering 30% population of the US,(34) with poor representation of the variable incidence and prognosis of EC worldwide.(35) Secondly, the qualified cases involved in our study was not enough to develop a strong nomogram model with considerable net benefit. Thirdly, some important information such as the life of quality, complications and treatment protocols were absent in the SEER database, therefore, the further analysis of certain specific subgroup could not be performed. Last, due to the lack of specific time-point for treatment, a clear causal relationship between treatment and prognosis could not be showed.