Clinicopathological data for metachronous disease
Out of the 346 primarily operated patients, 111 patients met our pre-defined inclusion criteria for analysis of isolated resectable metachronous disease during follow-up investigations (Table 1). In 102 patients, the primary PDAC was located in the pancreas head (91.9%) while in nine patients (8.1%) the tumor originated from the pancreatic tail (Table 1). In our total cohort of patients, the mean follow-up period was 10.1 months (range: 1-102.1 months). The median age of all 111 patients at the time of primary surgery was 69 years (range 41–95 years). Our collective consisted of 64 males (57.7%) and 47 females (42.3%). In correlation analysis of TNM staging of the primary tumor and site of metachronous disease, T- and N-stage, tumor grading, positive perineural and lymphatic invasion did not correlate with site of metachronous disease. However, patients with resected synchronous hepatic metastases (M1) and patients with positive venous invasion (V1) rater succumbed to metachronous hepatic metastases. (Table 1).
Surgical therapy and chemotherapy
Out of the total study collective, 31 patients received therapy for isolated local recurrence (group surglocal n = 11; chemlocal n = 20). Further 25 patients were treated for isolated metachronous pulmonary metastases (group surgpul n = 11; chempul n = 14) and 55 patients received therapy for isolated metachronous hepatic metastases (group surghep n = 6; chemhep n = 49) (Table 2).
For hepatic metastases, one right hemi-hepatectomy and five atypical non-anatomical resections were performed. For pulmonary metastases, five right and six left atypical resections via video-assisted thoracoscopy (VATS) were done. For local recurrence, one gastrectomy with atypical resection of the left diaphragm, nine salvage-pancreatectomies with simultaneous right hemicolectomies and one partial psoas muscle resection were performed (Fig. 1, Table 2). In all 28 patients, margin negative resections were achieved. Out of the 83 conservatively treated patients, 45 patients (54.2%) received gemcitabine mono, while a combination therapy with gemcitabine was administered in 19 patients (22.9%). In 5 patients (6.0%), FOLFIRINOX was applied.
Disease free survival and relapse specific survival
Median survival data is summarized in Table 3. The median disease free survival was with 18.15 months longer in patients before diagnosed pulmonary metastases when compared to patients before diagnosed isolated local recurrence or metachronous hepatic metastases (p = 0.029) (Table 3, Fig. 2A). The median DFS with 8.52 months before diagnosed local recurrence was similar to the median DFS with 6.77 months in patients before diagnosed hepatic metastases (p = 0.180) (Table 3, Fig. 2A).
In order to exclude selection bias and differences in tumor biology between the applied treatment modalities for metachronous disease, DFS before secondary relapse treatment initiation was stratified according to the treatment modality applied (Table 3). The median DFS before secondary therapy initiation was similar in patients with isolated hepatic metastases or local recurrence between both treatment modalities (surgery and chemotherapy) (p = 0.927 for surghep vs. chemhep and p = 0.666 for surglocal vs. chemlocal) (Table 3). However, DFS in patients before surgery for isolated pulmonary metastases was significantly longer when compared to patients who only received chemotherapy (p = 0.004) (Table 3). Hence a selection bias was detected only in patients with isolated pulmonary metastases.
To further elucidate a survival benefit between surgery and conservative treatment, the relapse specific survival for each analyzed relapse compartment was evaluated after therapy initiation. Surgical therapy significantly improved survival in patients with isolated local recurrence or pulmonary metastases, when compared to chemotherapy (p = 0.002 for surgpul vs. chempul; p = 0.011 for surglocal vs. chemlocal and p = 0.545 for surghep vs. chemhep) (Fig. 2B-D). Although the median RSS was similar in patients after diagnosed isolated pulmonary metastases and in patients after diagnosed local recurrence, a true cure from disease however was not established in patients with isolated pulmonary metastases as all patients succumbed after secondary therapy in our study (Table 3, Fig. 2B). Five from the 11 patients after surgically resected isolated local recurrence were still alive in our study 102 months after relapse surgery. In spite of a dismal median DFS in patients before diagnosed local recurrence, which was similar to patients with diagnosed hepatic metastases, long-term survivors (> 5 years) were found only in patients after surgically resected isolated local recurrence (Fig. 2D).