br Statistical analysis br The
The statistical analyses were performed with IBM SPSS (version 25, IBM Corp., Armonk, NY, USA). Normality was tested with the ShapiroeWilk test. Normally distributed continuous variables were assessed by Student's t-test and reported as the mean ± standard deviation. Categorical variables were evaluated with a Chi squared or Fisher's exact test. Otherwise, the ManneWhitney rank-sum test was used. Different treatment strategies and their DFS, Interval and OS were shown with a Kaplan-Meier survival curve, and a log-rank test was applied to compare these curves.
There were 80 patients included in this FilipinComplex study. The intervention group (A) consisted of 26 (32.5%) patients who started their salvage treatment at the time of significantly increasing CA19-9 and negative radiological examinations. The control group (B þ C) had 54 (67.5%) patients who were treated conventionally after metas-tases were confirmed by radiological examinations. The median follow-up time was 18.7 months. At the last follow-up, death was documented in 44 patients (55%) and 22 patients (27.5%) were still alive after a minimum follow-up of 12 months. Clinical character-istics of these patients are shown in Table 1. There was no
Baseline clinical characteristics of patients.
Control group P value
Adjuvant chemotherapy (n)
Recurrence site (n)
Salvage treatment (n)
Gemcitabine þ S-1 6 3 4
Chemoradiotherapy 2 9 3
a SOX: S-1 plus oxaliplatin; XELOX: capecitabine plus oxaliplatin.
statistically significant difference in prognostic factors between these two groups.
Most patients received single agent gemcitabine or S-1 as adjuvant therapy. Other additional treatment during adjuvant chemotherapy included chemoradiotherapy, oxaliplatin, irinotecan and paclitaxel-albumin. When patients started salvage treatment, most of them received gemcitabine or fluoropyrimidine based chemotherapy, which was different from their former regimen to overcome the drug resistance.
Rising tumor markers and radiological examination
The increasing tumor markers preceded the appearance of radiological signs of relapse in 60 of the 80 (75%) patients evalu-ated. In the remaining patients, no increase in the tumor marker was found at the time of relapse or the increased markers was contemporaneous with the radiological signs of metastases. The median interval between CA 19-9 elevation and radiological evi-dence of recurrence was 3.4 months.
Patients who started salvage treatment only based on rising CA19-9 levels had a significantly longer median DFS than patients with treatment changes based on radiological examinations (23.6 months vs. 12.1 months, P < 0.001) (Fig. 1A). Compared with the A and B group, the interval from CA19-9 elevation to radiological confirmation was also significantly prolonged in patients with tu-mor marker guided salvage treatment (13.2 months vs. 3.5 months, P < 0.001) (Fig. 1B). Apart from the longer DFS, the median OS was prolonged in the intervention group (28.1 months vs. 20.7 months, P ¼ 0.049) (Fig. 1C).
According to the ASCO guidelines, rising CA19-9 levels usually precede the radiographic appearance of a recurrent disease, but the present data are insufficient to recommend the routine use of CA19-9 alone for monitoring responses to therapy without radio-graphic confirmation . However, many studies have shown that there is a correlation between the patient's clinical benefits and tumor marker CA19-9 levels decline during chemotherapy for advanced pancreatic cancer, which means rising CA19-9 levels may serve as a negative predictive marker [14,26,27]. A recent study found that the time from CA19-9 elevation to radiographic recur-rence ranged from 6 to 18 months . No matter the type of cancer, the primary goal of surveillance after curative treatment is to detect the local or distant recurrence as early as possible and
doctors can intervene to prolong the survival of patients. Therefore, rising CA19-9 levels during adjuvant chemotherapy might be helpful to consider salvage therapy, especially when the radio-graphic evidence of tumor recurrence is absent. However, there is still not enough evidence to support the apparent common practice that regular follow-up after surgery to identify earlier recurrence can improve the survival of patients [28e32]. Although the role of surveillance in patients with resected pancreatic adenocarcinoma is limited, CA19-9 measurement and follow-up CT scans with contrast every 3e6 months for 2 years after surgical resection are still category 2B recommendations according to NCCN guidelines.