This study was an open-label, fast-track, randomized controlled, feasibility trial comparing the effect of CART vs. paracentesis alone for patients with refractory malignant ascites.29,30 This study was conducted from February 19th, 2018 to July 13th, 2020.
This study was conducted in accordance with the ethical standards of the Helsinki Declaration and the ethical guidelines for epidemiologic research of the Ministry of Health, Labour and Welfare of Japan. The protocol, procedures, information sheets, consent forms, and questionnaires were approved by the Institutional Review Board of each site. Patients gave written informed consent before enrollment. This study was registered at University Hospital Medical Information Network Clinical Trial Registry (UMIN-CTR) as UMIN000031029. CART was reimbursed by national health insurance.
Participants
This study recruited participants from two palliative care units and two hospital palliative care teams of four designated cancer hospitals. Inclusion criteria were: aged 20 years or older with diagnosis of metastatic or locally advanced cancer with malignant ascites, received paracentesis within 3 weeks and considered that repeated paracentesis would necessary for symptom control, admitted to participating institutes, and received no further anticancer treatments. Exclusion criteria were: life expectancy of two weeks or less, received chemotherapy, immunotherapy, or abdominal radiotherapy within 28 days, systemic or intra-abdominal infection, fever higher than 38 degrees Celsius, systolic blood pressure <= 80 mmHg, serum hemoglobin level < 6 mg/dL, serum creatinine level >= 2.0 ug/dL, serum total bilirubin >= 5.0 mg/dL, past history of hepatic encephalopathy, cognitive impairment, and no peripheral venous route.
Intervention
After enrollment, patients received allocated treatment: CART or simple paracentesis, when clinically indicated, i.e., driven mainly by worsening of the patient’s symptoms due to ascites accumulation. In both arms, ascites was removed under local anesthesia and mobilized by gravity. Drainage volumes of patients in each arm were decided by the primary responsible palliative care physician according to the national clinical guideline in which experts proposed 1 to 3 L of paracentesis as a balanced procedure.3
Following the removal of ascitic fluid, patients in the CART arm received reinfusion of removed ascitic fluid. After ascites drainage, removed ascitic fluid was filtered and concentrated using AHF-MOW and AHF-UP (Asahi Kasei Medical, Tokyo, Japan), which remove bacteria and malignant cells and concentrate materials with a molecular weight of more than 30,000 including autologous albumin. The filtered and concentrated ascetic fluid was reinfused into the patient via a peripheral vein at 50 – 100 mL per hour. We injected 400 mg of acetaminophen intravenously before reinfusion for the prophylaxis of fever.
Patients in the control (waiting list) arm received CART after the study when the next paracentesis was needed.
Endpoints
Our primary objective was to determine the feasibility of conducting this randomized trial to compare CART vs. paracentesis. The primary feasibility endpoint was the percentage of the patients who completed the study intervention among those patients who consented.
The secondary endpoints included paracentesis-free survival, patient request on the questionnaire for paracentesis (PRO-paracentesis)-free survival, and adverse events. Paracentesis-free survival was defined as the time from the day of allocated treatment (day 1) to the day of the next paracentesis or death, whichever occurred first. PRO-paracentesis-free survival was a patient-reported outcome measure defined as the time from day 1 to the day when the patient first reported on the questionnaire that they would want a paracentesis if indicated, received the next paracentesis, or died, whichever occurred first.
We further explored multiple potential outcome measures of ascites treatments: the changes of patient reported physical symptom intensities, i.e., the total score of the Edmonton Symptom Assessment System: Ascites Modification (ESAS:AM) and a separate single question of abdominal distension numerical rating scale (NRS), and the change of abdominal circumference.
Outcome measures
Patient’s request for paracentesis on the questionnaire: Patients answered yes or no to the question “If indicated, would you want a paracentesis now?” on the questionnaire.
Abdominal distension numerical rating scale (NRS): Patients rated the current intensity of abdominal distension using an 11-point numerical rating scale, ranging from 0 (none) to 10 (worst possible).
Edmonton Symptom Assessment System: Ascites Modification (ESAS:AM): ESAS:AM includes 11 items that consist of nine items of the original ESAS and two additional ascites-related items (‘‘abdominal distension’’ and ‘‘mobility’’).7,31 Patients rated the intensity of each symptom at a point in time (current) using an 11-point numerical rating scale, ranging from 0 (symptom absent or best) to 10 (worst possible). The validity and reliability of the Japanese version of the ESAS:AM were previously confirmed.32
Abdominal circumference: Abdominal circumference was measured by the attending nurse of the day at the level of the umbilicus with the patient in a supine position.33
Procedures
We followed patients from the day of the allocated treatment (day 1) to the day of the next paracentesis or day 14, whichever occurred first. Outcome measurement and analyses were done for the allocated treatment only.
At enrollment, we obtained data about patient demographics such as age, sex, ECOG performance status, primary tumor site, presence of metastasis in the liver and lung, co-treatments (opioid, diuretics, and artificial hydration), serum laboratory data investigated within a week (total protein, albumin, total bilirubin, and creatinine), and ascites laboratory data investigated within 8 weeks (red blood cell count and albumin). Ascites was defined as hemorrhagic if the ascitic red blood cell count was more than 10,000/μL.34
During the 14-day follow-up period, patients filled in the questionnaire on their request for paracentesis and abdominal distension NRS once a day, and ESAS:AM on days 1, 2, 4, 6, 8 at 10 to 12 o’clock. Nurse or family caregiver assistance was allowed in the case that patients found it difficult to fill in the questionnaire by themselves (proxy rating was not allowed). Abdominal circumference was measured once a day at 10 to 12 o’clock.
On day 14, primary responsible palliative care physician recorded the day of the next paracentesis or death during the observation period. Known adverse events, i.e., fever, hypotension, bleeding, and hepatic encephalopathy, possibly caused by the interventions were recorded according to Common Terminology Criteria for Adverse Events (CTCAE) version 4.0.
Sample size
We planned to enroll 10 patients per arm, with which we would be able to estimate the percentage of patients who completed the study intervention of 80% to within a 95% confidence interval of 62 and 98%.35,36 This study was not powered for a direct comparison between CART and paracentesis.
Randomization and blinding
Patients were randomly assigned in a 1:1 ratio to either the CART (fast-track) arm or control (waiting list: simple paracentesis) arm. Allocation was done by minimization to balance two potential confounders: hemorrhagic ascites (absent or present) and serum total protein (<5.5, >=5.5 g/dL).25 All patients, clinicians, and investigators were aware of the treatment allocation. All clinicians and investigators were blinded to the questionnaire, except for the attending nurse who supported the patients to fill it in.
Statistical analysis
We summarized the baseline demographics and feasibility outcomes using descriptive statistics. Then, Kaplan-Meier curves were computed for paracentesis-free survival and PRO-paracentesis-free survival following the intention to treat principle. We described adverse events of each arm. As a feasibility trial, these endpoints were summarized descriptively, with no formal statistical comparisons between arms.
Next, we tested the differences in the abdominal distension NRS score, total score of the ESAS:AM, and abdominal circumference of patients who completed the observation at 3 time points: day 1, day 2, and the day of the next paracentesis (or the day closest to the next paracentesis day of the days when ESAS:AM was measured), using one-way repeated measures ANOVA. As we explored whether these measures reflected the effect of ascites drainage and the recurrence of ascites, we did not perform a statistical comparison between arms.
Then, we visualized the changes in outcome measures for each patient, and classified them into stable increase and random changes. We assessed the agreement between the day of receiving the next paracentesis and the first day of the patient’s request for paracentesis: agreement with the request: received paracentesis within 2 days of the first request, delayed paracentesis: received paracentesis 2 days or more after the first request, early paracentesis: received paracentesis before the request.
All statistical analyses were performed using R, version 3.5.3 (R Core Team 2019, Vienna, Austria). All P-values were 2-sided. A P-value of <0.05 was considered significant. Survival analyses were performed using the ‘‘survival’’ package.