The median follow-up period was 21 (range: 0-121) months. In the entire group of patients, the median OS was 34 months, and the 3- and 5-year survival rates were 48.5% and 36.9%, respectively.
Among the 134 patients undergoing pancreas resections for PDAC, 46 patients (34%) were aged 75 years or older. The comparison of clinicopathologic characteristics and postoperative outcomes between the elderly and younger patients was shown in Table 1. Although the elderly patients had more risk and trend with aspect to hemoglobin level and NLR (P = 0.010, P = 0.062), there were no significant differences in the induction and completion rates of NACRT, intraoperative variables including the surgical procedures, incidence of portal vein resection, length of operation and blood loss, and pathological data of LN metastasis and resection status between the two groups.
Mortality occurred in a patient in the younger group. Morbidity [18] (≥ Grade IIIb) occurred in 7 (15%) in the elderly group. The elderly group experienced a significantly higher incidence of postoperative major complications (P = 0.046). The incidence of POPF was similar between the two groups, while DGE occurred more frequently in the elderly patients than in the younger patients (P = 0.001). Furthermore, postoperative hospital stay in
the elderly group was longer than that in the younger group (P = 0.008). The induction rate of ACT was similar between the two groups (74% vs 80%, P = 0.458), however, the completion (more than 6 months) rate was significantly lower in the elderly group than the younger group (P = 0.022). Only 16 (35%) patients could complete ACT in the elderly group.
NACRT was performed for 82 patients (61%). Twenty-seven (59%) patients in the elderly group received NACRT (2 weeks: n = 18 and 5 weeks: n = 9), and 55 (63%) in the younger group did (2 weeks: n = 36 and 5 weeks: n = 19). There were no significant differences about the induction of NACRT between the two groups (P = 0.668). As for the completion rate, 24 out of 27 (89%) completed NACRT protocol in the elderly group and 48 out of 55 (87%) in the younger group. There were no significant differences about the completion rate of NACRT between the two groups. More importantly, both groups had high completion rate of NACRT regardless of the treatment period.
The median overall survival (OS) time and 3-year OS rate in the elderly patients were 27 months and 45%, compared to 58 months and 52% in the younger patients, respectively (Fig. 1a). As well, the median relapse free survival time and 3-year relapse free survival rate (RFS) in the elderly patients were 18 months and 36%, compared to 16 months and 36% in the younger patients, respectively (Fig. 1b). The differences were not statistically significant both in OS and RFS between the two groups (P = 0.270, P = 0.699).
Table 2 shows the subgroup analysis for the clinicopathological factors in elderly patients with (n = 27) and without (n = 19) NACRT. The background factors such as age, sex, BMI, resectability [15], serum albumin and hemoglobin, NLR, intraoperative blood loss, transfusion, pathological LN metastases and resection status were not significantly different between the patients with and without NACRT. PD and TP were more frequently performed in patients with NACRT (P = 0.014). Portal vein was more frequently resected in the NACRT group (P = 0.044), and operation time was significantly longer in patients with NACRT (P = 0.049). As for perioperative outcomes, there were no significantly differences in morbidity, postoperative hospital stay and ACT induction and completion rates between the patients with and without NACRT.
The elderly patients with completion of postoperative ACT had significantly better OS than those without completion or induction of ACT (P = 0.032) (Fig. 2). The median OS and 3-year OS rate in the elderly patients with completion of ACT were 45 months and 66%, compared to 18 months and 33% in its counterpart. On the other hand, NACRT itself did not significantly affect OS in elderly patients, however, there was a trend toward improvement of OS (P = 0.072) (Fig. 3).
Figure 4 showed OS curves in the elderly group dividing into 4 groups with and without inductions of NACRT and ACT. Only patients who introduced both NACRT and ACT had significantly better OS than those who did either of NACRT or ACT or did not do either (P = 0.042, P = 0.017 and P = 0.002). It was suggested that performing NACRT before surgery and further performing ACT after surgery would improve OS for the elderly patients.