Older adults more commonly suffer from esophageal squamous cell carcinoma (ESCC), and as populations age, the average age of those affected also increases (1). Comorbidities and critical dysfunction in pulmonary, cardiac, or renal organs, for example, are often present in older patients (2), who cannot tolerate treatment intensity as easily as younger patients. However, as the clinical data clearly show statistically, surgery alone cannot control advanced ESCC (3). In Japan, esophageal cancer in the surgically resectable stages is generally treated with neoadjuvant chemotherapy (NAC) and subsequent surgery (4,5). Recent results from the JCOG1109 randomized clinical study have changed the standard treatment for patients with clinical stage II or III ESCC in Japan. Now, neoadjuvant triplet chemotherapy with docetaxel, cisplatin, and 5-fluorouracil (DCF) is administered in place of cisplatin (CDDP) plus 5-fluorouracil (5-FU) (6,7). However, the JCOG1109 study included only patients aged ≤75 years with an Eastern Cooperative Oncology Group performance status (PS) of 0 or 1 (8). Deciding how to treat older patients based on clinical trial results from younger patients can be difficult in real-world clinical practice. By dividing the doses of docetaxel (TXT), CDDP, and 5-FU, new regimens with high completion rates and therapeutic efficacy are being developed (9–11). Identification of the increasing number of older patients with ESCC who are intolerant to preoperative treatment vs. those who should be treated preoperatively could speed the development of appropriate therapeutic strategies.

We therefore conducted a multicenter retrospective study to determine the indications for divided-dose DCF (biweekly DCF) in patients aged ≥70 years with ESCC in comparison to upfront surgery (US).

Surgery is a particularly invasive treatment for ESCC. Nevertheless, it has remained the primary form of treatment for locally advanced ESCC even though perioperative treatment has intensified and improved the prognosis. Recent advances have increased the safety of surgical treatment, and more facilities are actively performing surgery on older patients with ESCC (17). In a study comparing 50 esophageal cancer patients ≥75 years old with 100 patients <75 years old, Kanda et al (18) reported no significant differences in postoperative complications. Morita et al (19) reported a morbidity rate of 25% for esophagectomy in patients ≥80 years old and found the incidences of surgical and medical complications to be similar to those for patients <70 years old. Moreover, they reported a decreased morbidity rate even in their patients >80 years old by following strict indications for surgery and performing a less invasive operation (omitting supraclavicular lymphadenectomy and performing a two-stage operation for risky patients). In their study of 5,066 patients aged 75–79 years old with ESCC, Motoyama et al (20) reported that surgery significantly prolonged OS compared to chemoradiation therapy or chemotherapy alone in advanced esophageal cancer of stage II or higher. In contrast, Miyata et al (21) reported that among 722 esophageal cancer patients >70 years old divided into four groups according to age, respiratory and cardiac complications increased with age. Older patients are particularly faced with many age-specific problems, such as aspiration pneumonia from delayed recovery of swallowing function, prolonged hospitalization due to decreased activities of daily living, and even progression of dementia.

There are several reports on the benefits of NAC to treat esophageal cancer in older patients. Yamashita et al (22) compared data on patients aged ≥75 years with advanced ESCC receiving NAC or not and found a better prognosis in those patients responding pathologically to NAC. However, in their patients with a PS of 1 or higher, the prognostic value of NAC was not clear, and they suggested that this group could likely undergo surgery alone. Among older patients with ESCC and a poor PS, Booka et al (23) found NAC to be non-beneficial and considered an increase in postoperative complications as the reason for NAC worsening the prognosis of these patients. Matsuda et al (24) similarly reported no survival benefit with preoperative DCF, the current standard of treatment, in patients >76 years old. Furthermore, they reported that pneumonia and anastomotic leakage as postoperative complications were negative prognostic factors for shorter OS and RFS in patients with esophageal cancer who were >75 years old and had undergone preoperative therapy with DCF (25).

Myelosuppression may be reduced by the divided administration of TXT and CDDP without greatly changing its efficacy (10). Neutropenia was the most common Grade 3 or higher toxicity in 31.3% of the patients in the biweekly treatment regimen, whereas Kato et al (7) reported that 85% of their patients developed Grade 3 or higher neutropenia. In the present study, we limited the NAC regimen to biweekly DCF. Although this regimen was reported to be a less toxic and potentially effective treatment, it did not show usefulness as NAC in an older population (9,10). This result is similar to and supports that reported in the previous literature (22,23,25). Although there is no difference in long-term prognosis, it may be better for older patients with ESCC to undergo US to avoid the side effects and decreased physical strength resulting from NAC. In our examination of surgical outcomes, the incidence of failure resulting in anastomotic leakage was different between the NAC group and US group. This was presumably due to differences in fine anastomotic technique and gastric tube construction between centers.

In NAC for ESCC, the PMI has a significant effect on differences in chemotherapy response rates and adverse event rates (11,14,26,27). Our cohort showed significantly prolonged 3-year OS in the PMI low group of the NAC group compared to that in the US group. The usual duration of NAC of eight weeks or more is an active period of nutritional management and intervention with rehabilitation. The present results suggest that for older patients with ESCC and low PMI, the duration of NAC may also lead to a period of careful preoperative preparation, which may result in a favorable outcome by selecting eligible patients for surgery. In fact, the PMI low group tended to have higher PMI due to multifaceted therapeutic interventions during the NAC (Fig. 5). However, as low PMI itself is a favorable factor for adverse events, it is important to perform NAC safely and in conjunction with the delivery of adequate nutritional therapy and rehabilitation that maintains muscle mass. It is possible that the positive impact of lower toxicity by dividing DCF into a biweekly regimen had an oncological effect in the low PMI group.

This study has several limitations. First, selection bias was likely present due to the retrospective nature of the study. Second, although this study focused only on a treatment regimen of biweekly DCF, dose intensities were not analyzed. Further, patients unable to undergo esophagectomy with no known reason for discontinuation, such as disease progression or toxicity during NAC, were excluded. Third, limited information was collected about patient characteristics, and preoperative pulmonary function or other factors were not evaluated. No power calculations were performed in the PMI study because recruitment was opportunistic. Fourth, consensus on the indications for postoperative adjuvant therapy in the US group was lacking. Fifth, there was a relatively short observation period.

We found that compared to US, a biweekly DCF treatment regimen did not prolong OS and RFS at all stages in patients with advanced ESCC who were ≥70 years old. Further prospective large-scale studies will be required to develop an optimal treatment strategy that is less toxic to but maintains efficacy in older patients with advanced ESCC.