This is a retrospective observational cohort study. We reviewed the medical records of 454 patients with unresectable pancreatic cancer who were treated between 1987 and 2017. The median age of all patients was 61 years (range 18–80). There were 192 female and 262 male patients; 391 patients had adenocarcinoma; 7 patients had endocrine carcinoma and 56 patients had an unknown histological status. Regarding classification, 174 patients were classified stage III, and 280 patients were classified stage IV. Furthermore, 265 patients had no prior treatment, and 189 patients had prior treatment with systemic chemotherapy, chemoembolization, and/or irradiation. Observation time was at least 17 months (Table 1).
Investigations were performed in compliance with the principles of good clinical practice outlined in the Declaration of Helsinki and federal guidelines and were approved by the Medias Institutional Review Committee. Informed consent was obtained from each participant or participant’s guardian.
The mode of drug administration was either intra-arterial infusion through an angiographic catheter alone (233 patients) or a combination of intra-arterial infusion and at least one upper abdominal hypoxic perfusion with chemofiltration (UAP-F; 221 patients).
Technique of upper abdominal hypoxic perfusion with chemofiltration (UAP-F)
Upper abdominal perfusion is a method performed in two steps, where the first step is the stop-flow procedure and the second step is the isolated hypoxic abdominal perfusion (Fig. 1a). Both steps are performed with stop-flow balloon catheters that are first inserted through the femoral artery and vein in the groin area and threaded into the vena cava and aorta. The venous balloon is placed beneath the diaphragm. Its position is controlled through radiography by temporarily inflating it with contrast medium, followed by deflation until the therapy is started.
Stop-flow balloon catheters consist of three channels: one for the isolated perfusion, one for inflation of the balloon, and one coaxial channel exiting at the tip above the balloon for potential insertion of a guidewire and/or injection of chemotherapeutics as well as monitoring of arterial blood pressure during the procedure (Fig. 1b).
For the first step (stop flow), the balloon in the aorta is positioned right beneath the celiac trunk and inflated, and its correct position is controlled by radiography with intra-arterial contrast medium. Chemotherapy is infused for 1 min under high oxygenation while an outflow-block of the liver veins is contemporarily established by inflating the venous balloon. Because of the infra-celiac aortic balloon block, the first-pass flow of chemotherapeutics exits exclusively through the celiac trunk that supplies the tumor region. Thereafter, the aortic balloon is immediately slipped upstream in the aorta and placed right beneath the diaphragm. At this point, the hypoxic part of the procedure begins. Thus, both balloons stop the blood flow, and a high drug concentration is maintained in the distribution area of the celiac trunk in the upper abdominal region for 5 min.
During the second step, the isolated hypoxic abdominal perfusion is started via the side holes in the perfusion channel of the catheter beneath each balloon. After 5 min of perfusion with the high drug concentrations in the whole abdominal region, another 5 min of perfusion with additional chemofiltration is maintained in order to lower the cytostatic load before deflating the balloons and continuing chemofiltration until a substitution volume of 4 L is reached (Fig. 1a).
The isolated hypoxic abdominal perfusion (HAP) step has been applied 520 times for 221 patients, upon which 318 procedures comprised the stop-flow interval and HAP, and 202 procedures were application of HAP only.
Intra-arterial infusion chemotherapy via angiographic catheters in the celiac trunk
Intra-arterial infusion chemotherapy is performed in local anesthesia via a Sidewinder angiocatheter. It has been administered 977 times for 233 patients; in 26 instances, additional chemofiltration was applied, and in 231 procedures, it was combined with chemoembolization with starch microspheres.
Treatment cycles and drug regimen
Drug combinations for UAP and HAP were 50 mg cisplatin, 30 mg adriamycin and 15 mg mitomycin C for an average 70 kg patient. Adriamycin and mitomycin C have been shown to have a particular strong cytotoxic effect under hypoxic conditions while cisplatin works equally well under aerobic and hypoxic conditions (Teicher et al. 1981). For i.a. chemotherapy infusion via an angiocatheter, a drug combination of 30 mg cisplatin, 2 × 15 mg adriamycin and 10 mg mitomycin C was administered for 5 min each on four consecutive days. In addition, 29 patients also received cycles with 600–1000 mg gemcitabine and cisplatin. Depending on disease extension and response, 2–10 treatment cycles in 3 weeks intervals each were administered. The median number of cycles was E4. One exceptional patient received 26 cycles over a 6-year period.
Blood sampling methods for cisplatin plasma concentration measurements
A series of cisplatin plasma concentration measurements has been investigated. The measurements were obtained during and after intra-arterial infusion of CDDP (cisplatin) through the coaxial channel of the balloon catheter. For intra-arterial infusion, the duration was chosen according to the infusion duration in tumors of the head and neck, which was 5–7 min (Aigner et al. 2018, 2019). Its concentrations were measured at minutes 1, 2 and 3, respectively, with measurements collected at 2-min intervals. The arterial blood samples were taken with a Sidewinder-II catheter from the common hepatic artery supplying the tumor. The venous blood samples were taken from a central venous catheter. The blood samples were centrifuged and plasma drug concentrations were measured.
Criteria for response and adverse events
Tumor responses were assessed in accordance with Response Evaluation Criteria in Solid Tumors (RECIST version 1.1) at 2 weeks after every second treatment cycle. Responses were evaluated by computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET). Pain controlled by < 50% analgesic administration at 20 days after treatment was considered objective pain relief. Adverse events were assessed according to the common terminology criteria for adverse events of the National Cancer Institute.
Statistics were calculated with 95% confidence limits. Survival times were estimated using the Kaplan–Meier product limit estimator, and follow-up for surviving patients was minimum 10 months, with a median follow-up of 39 months. Survival times were stratified according to clinical variables that may affect survival, and log-rank tests were used to verify significance. Statistical analyses were performed by using MediasStat software version 28.5.14.