The predictors and patterns of the early recurrence of pancreatic ductal adenocarcinoma after pancreatectomy: the influence of pre- and post- operative adjuvant therapy

Pancreatic ductal adenocarcinoma (PDAC) has a poor prognosis; only 3% of patients survive at 5 years after diagnosis [1, 2]. Only 20% of patients with PDAC are eligible to undergo radical resection [3]. Although surgical resection provides the only chance for a cure, it is associated with a median overall survival (OS) period of 11 to 23 months, with a 5-year OS rate of about 20% [4, 5].

Wagner M et al. reported that R0 resection is the most important factor determining outcome in patients with PDAC [6], and Ihsan ED et al. reported that R1 was associated with a decreased OS and disease free survival (DFS) in PDAC when compared with R0 [7]. The early recurrence of PDAC postoperatively is a frequently observed, serious problem, even after microscopically curative resection is performed. La Torre et al. reported that 60% of patients experience local or systemic recurrence within the first 12 months after curative resection [4]. Some reports suggested that the preoperative factors that are associated with the survival time after surgery were tumor size [4], preoperative lymph node metastasis [4], the preoperative serum carbohydrate antigen 19–9 (CA19–9) level [3, 4, 8, 9], histological grades [4, 9], duration of symptoms [3], and the preoperative modified Glasgow Prognostic Score [10]. These might be predictors of the early recurrence of PDAC postoperatively.

Neoadjuvant therapy was not actually recommended for patients with resectable (R)-PDAC in the 2016 National Comprehensive Cancer Network (NCCN) guideline [11], but neoadjuvant chemotherapy or chemoradiotherapy (CRT) may reduce the early recurrence of PDAC. Upfront surgery might be a predictor of the early recurrence of PDAC, even for those with R-PDAC.

The aim of this study was to detect factors influencing on early recurrence and its patterns for the patients with PDAC including neoadjuvant CRT and adjuvant chemotherapy (ACT).

This study was approved by the institutional review board of Kagawa University. A total of 142 consecutive patients undergoing pancreatectomy for PDAC between January 2000 and May 2016 were retrospectively examined. Informed consent was obtained from all patients according to the institutional protocol of our hospital. All 142 patients had a PDAC that was histologically examined by at least two pathologists. Of the 142 patients, 27 patients were excluded. Ten patients were censored within 6 months, 10 were classified as unresectable category based on the 2016 NCCN guideline [11], 5 had unclear recurrence timing, 1 underwent R2 resection, and 1 had perioperative mortality. The data from the remaining 115 patients were retrospectively analyzed.

The patients were diagnosed with R (n = 86) or borderline resectable ([BR], n = 29) PDAC according to the 2016 NCCN guidelines [11]. All surgical procedures were divided into the following three types: classic, pylorus-preserving, or subtotal stomach-preserving pancreaticoduodenectomy (PD) in 75 patients (65%); distal pancreatectomy in 36 (31%); and total pancreatectomy in 4 (3%). Systematic lymph node dissection was performed in all operations. R0 resection was achieved in 104 patients (90%) and R1 was achieved in 11 (10%). R0 resection was defined as negative margin based on Union for International Cancer Control (UICC) definition. Preoperative CRT and postoperative ACT were given to 51 (44%) and 85 (74%) patients, respectively. Among the 115 patients, 34 (30%) experienced early recurrence within 6 months postoperatively (group E) and 81 (70%) did not (group NE).

The median follow-up period was 18 (range: 2–112) months. In the entire group of patients, the median OS period was 24 months, and the 3- and 5-year survival rates were 45.9 and 28.7%, respectively. The median OS times of the groups E (n = 34) and NE (n = 81) were 9 and 37 months, respectively (P < 0.001).

Table 1 shows the results of the univariate and multivariate analyses of the factors that possibly affected the early recurrence of PDAC among the 115 patients. Significant associations with early recurrence were observed for the serum CA19–9 level ≧173 (P = 0.026), NLR ≧4.65 (P = 0.020), lymphocyte count ≦1648 (P = 0.042), SUV on FDG-PET ≧4.73 (P = 0.043), postoperative serum CA19–9 de-normalization defined as no return into the normal range (≥37 U/ml) except in patients who had normal preoperative CA19–9 (P < 0.001), no postoperative ACT (P < 0.001), pathological tumor size ≧3.0 cm (P = 0.028), para-aortic lymph node metastases (P = 0.009), and positive serosal (S) and plexus (PL) invasion factors (P = 0.009 and P = 0.029, respectively). Preoperative CRT did not reach statistical significance in the univariate analysis (P = 0.093).

In the multivariate analysis of factors that were found to affect the early recurrence of PDAC in the univariate analysis, postoperative serum CA19–9 de-normalization (odds ratio [OR], 23.10; 95% confidence interval [CI], 4.21–126.86; P < 0.001), no postoperative ACT (OR, 10.41; 95% CI, 2.73–39.64; P = 0.001) and positive S factor (OR, 4.94; 95% Cl, 1.47–16.62; P = 0.010) were independent risk factors for the early recurrence of PDAC. CA19–9 denormalization was seen in 17 of 21 patients (81%) who had preoperative CA19–9 elevation.

Table 2 shows the subgroup analysis results of the factors affecting the early recurrence of PDAC in the 51 patients with preoperative CRT and in the 64 patients with upfront surgery. The percentages of background factors such as resectability (P = 0.720), surgical procedure (P = 0.756) and Union for International Cancer Control stage (P = 0.808) were not significantly different between the E and NE groups (data not shown). Postoperative serum CA19–9 de-normalization (OR, 83.36; 95% CI, 3.32–2095.03; P = 0.007) and a positive S factor (OR, 13.08; 95% Cl, 1.25–137.46; P = 0.032) were independent risk factors for the early recurrence of PDAC in the multivariate analysis. Postoperative ACT did not significantly affect the early recurrence of PDAC in the multivariate analysis of this subgroup. On the other hand, the results of the univariate and multivariate analyses of the factors affecting the early recurrence of PDAC in the 64 patients who underwent upfront surgery revealed that postoperative serum CA19–9 de-normalization (OR, 39.26; 95% CI, 3.65–422.10; P = 0.002) and no postoperative ACT (OR, 15.53; 95% CI, 2.51–96.04; P = 0.003) were independent risk factors for early recurrence in this subgroup.

Recently, the efficacy of preoperative neoadjuvant chemotherapy or CRT for PDAC has been reported, especially for patients with BR- or locally advanced unresectable-PDAC [16‐18]. However, neoadjuvant therapy is not recommended for patients with R-PDAC in the NCCN guideline [11], presumably because of insufficient evidence. Preoperative therapy, however, might have a beneficial effect on patient survival in those with R-PDAC through preventing the early recurrence that is often seen even after curative resection.

The postoperative recurrence patterns of PDAC in 34 patients in the group E regarding preoperative CRT and postoperative ACT are shown in Fig. 1. The recurrence patterns were defined as the location of the first recurrence. Local recurrence included regional lymph node and plexus nerve recurrence base on radiological findings. We made a final decision while taking account of comprehensive set of factors in terms of tumor marker, FDG-PET and MDCT. In addition, EUS-FNA was applied in several cases with difficult lesion for diagnosis. The recurrence patterns showed that 13 (38%) initially had local recurrence and 21 (62%) patients initially had distant metastasis. Early local recurrence occurred in only 2 patients among a total of 51 patients receiving CRT preoperatively. Even in a group with surgery plus preoperative CRT alone (n = 12), no early local recurrence occurred. In contrast, it was seen in 5 (28%) patients undergoing surgery alone (P = 0.046) and in 6 (13%) patients with surgery plus postoperative ACT alone. On the other hand, early distant recurrence developed in only 4 (9%) patients with surgery plus postoperative ACT alone. In sharp contrast, it occurred in 8 (44%) who underwent surgery alone (P = 0.004) and in 5 (41%) patients with surgery plus preoperative CRT alone. Furthermore, both early local recurrence and distant recurrence significantly decreased in patients who received combined preoperative CRT and postoperative ACT, compared with surgery alone (P = 0.015 and P = 0.004, respectively). These results clearly demonstrated that preoperative CRT strongly prevented local recurrence but not distant recurrence, and postoperative ACT prevented early distant recurrence but not local recurrence.

The potential advantages of the preoperative delivery of CRT include the ability to sterilize tissues at critical margins oncologically. In the current study, the exposure dose of preoperative CRT was relatively low, at a minimum value of 30 Gy. However, a report from the M.D. Anderson Cancer Center contended that hypofractionated CRT (30 Gy) was associated with margin-negative resection rates, treatment effects, local control, and OS, similar to those associated with standard fractionated CRT (50.4 Gy) [19]. Similar results were proven in our previous study [13] and the current study.

This study demonstrated that the lack of postoperative ACT was a significant predictor of early recurrence. As supported by the results from previous clinical trials [14, 15], ACT is one of the most important factors for preventing early recurrence postoperatively. Judging from the early recurrence pattern that was found in this study, postoperative ACT prevents the early distant recurrence of PDAC, but it might have almost no effect on early local recurrence. Interestingly, there was no significant difference between patients with and without postoperative ACT in those with preoperative CRT. We speculated that preoperative CRT might compensate for a lack of postoperative ACT.

Preoperative CRT itself was not found to be an independent preventive factor against the early recurrence of PDAC in the univariate and multivariate analyses of the entire series of patients. However, the current study demonstrated that preoperative CRT significantly prevented local recurrence. Early local recurrence occurred in only 2 patients among 51 patients receiving CRT preoperatively. Based on these results, preoperative CRT possesses strong efficacy regarding local control and might prevent the early local recurrence of PDAC. In this series, 39 patients were treated with both preoperative CRT and postoperative ACT. In this subgroup, the early recurrence of PDAC occurred in only 6 patients (15%).

The current study has several limitations. This was a retrospective study that was conducted at a single institution. Therefore, the sample size was small and a historical backdrop existed. The types of preoperative examinations such as FDG-PET and postoperative adjuvant therapy changed during the study period. Within the past several years, preoperative CRT has especially been performed for patients with R- and BR-PDAC. Furthermore, there are missing values for several factors in the tables, and the AUCs in the ROC were relatively low and might indicate inadequate cutoff points.

CA19–9 de-normalization was an important predictor of the early recurrence of PDAC within 6 months after pancreatic resection. Postoperative ACT was an important preventive measure for the early recurrence of PDAC, particularly for distant recurrence. Preoperative CRT had a strong potential to prevent the early local recurrence of PDAC. In addition, preoperative CRT might compensate for the lack of postoperative ACT. In patients who are not expected to be capable of receiving postoperative ACT, preoperative CRT should be considered.