Efficacy of the peritoneovenous shunt as palliative care for cancer patients with refractory ascites: a randomized controlled trial (JIVROSG-0803 study)

doi:10.21203/rs.3.rs-2730903/v1

Download PDF

Research Article

Efficacy of the peritoneovenous shunt as palliative care for cancer patients with refractory ascites: a randomized controlled trial (JIVROSG-0803 study)

https://doi.org/10.21203/rs.3.rs-2730903/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

A randomized, controlled trial to evaluate the superiority of peritoneovenous shunt (PVS) over the other treatments for refractory ascites in advanced cancer patient was conducted.

Subjects and methods

The subjects were advanced cancer patients with symptoms due to refractory ascites which are difficult to control by medical therapy and cannot be improved by methods other than drainage and with a Palliative Prognostic Index of <6. They were randomly allocated in a 1:1 ratio to receive either PVS placement (test group) or the other treatment (control group). The area under the curve (AUC) of symptom improvement score curve calculated with visual analog scale was compared between the groups as the primary endpoint. The AUC of EQ-5D, SF-8 were also compared between the groups, and adverse events were evaluated with the CTCAE version 3 up to 4 weeks after enrollment or until protocol treatment was discontinued.

Results

Forty patients were enrolled between November 2009 and July 2018, with 20 allocated to test group and 20 control group. The mean AUC of the symptom improvement score curves were 133.2 (95% CI: 98.6-167.7) in the test group and 69.9 (95% CI: 34.4-105.4) in the control group (p=0.014). Therefore, the mean AUC was significantly larger in the test group, resulting the primary endpoint was met. There was no significant difference between the groups in SF-8 and EQ-5D. Though some known adverse events were observed, there was no significant difference in the frequency of individual adverse events between the groups.

Conclusions

PVS significantly improves the symptom of pain due to refractory ascites compared to other treatments. Although not as effective as improving the comprehensive QOL evaluated with SF-8 and EQ-5D, PVS may be considered the standard of care for advanced cancer patients with refractory ascites.

randomized controlled trial

peritoneovenous shunt

palliative care

cancer

refractory ascites

Medical treatment such as the administration of diuretics and albumin preparations is usually provided as the first choice for treating refractory ascites in patients with advanced cancer. However, such patients are often resistant to the treatment, and they experience severe satiety, thirst, and activity restriction, leading to deterioration of their quality of life (QOL) [1, 2]. Though the abdominal paracentesis is sometimes carried out to improve the symptoms, the efficacy is temporary. The peritoneovenous shunt (PVS) is recognized as one of the palliative treatments for refractory ascites, and the clinical effectiveness of PVS placement is reported as 50–80% (3–21). However, it is an invasive treatment, although it can be performed with local anesthesia, and various complications have been reported [3–21]. In order to clarify whether PVS truly contributes to the improvement of pain in patients with refractory ascites, it is necessary to compare it with other treatments from the patient’s perspective. Therefore, this randomized controlled trial was conducted to evaluate the superiority of PVS compared to the other treatments for refractory ascites in cancer patient.

This study was performed in line with the principles of the Declaration of Helsinki. The approval was granted by the Ethics Committee of National Cancer Center (2009/No. 2009-032). Also, this study was approved by the institutional review boards of all participating institutions. (trial registration number: UMIN-CTR 000003469).

Study design and patients

In this prospective multi-institutional randomized controlled trial, refractory ascites was defined as 1) apparent swelling of the abdominal wall caused by ascites on computed tomography images and 2) continuous improvement of symptoms could not be obtained by medical therapy except paracentesis.

The inclusion criteria were as follows: (1) patients with symptoms due to intractable ascites, which are difficult to control by medical therapy and cannot be improved by methods other than drainage; (2) Palliative Prognostic Index < 6 points; (3) expected survival period ≥ 4 weeks; and (4) written informed consent provided by the patient. The exclusion criteria were as follows: (1) patients who already received PVS; (2) those with intractable ascites occurring for 4 weeks or more; (3) those with bloody, mucous, and infectious ascites; (4) those with ascites due to malignant mesothelioma or liver cirrhosis; (5) those with clinically apparent heart failure, renal failure, and bleeding tendency; and (6) pregnancy or potentially pregnant women.

Eligible patients were randomly allocated 1:1 to the PVS group (test group) and control group using the minimization method with institution as an allocation adjustment factor. The protocol treatment for the test group was PVS placement. The protocol treatment for the control group was any kind of treatment except PVS placement. Termination of the protocol treatment in the test group was granted if the patient desired to undergo paracentesis because of their ascites-related symptoms or the physician deemed withdrawal from the study necessary because of adverse events. Termination of the protocol treatment in the control group was granted if the patient desired to undergo PVS placement. Treatments after termination were not regulated.

PVS placement

PVS placement was performed by interventional radiologists in an angiography suite. All procedures were performed with the patients under local anesthesia, and ultrasonography and fluoroscopy were used as the guiding modalities as previously reported [18–21]. The use of analgesics, sedatives, prophylactic antibiotics, steroids, and dopamine during the perioperative period was determined in accordance with the local practice of each participating hospital. The PVS system consisted of a 16-French (F) peritoneal catheter with side holes, 12-F venous catheter, and chamber that connected the two catheters. The chamber had a one-way valve, and it could be compressed manually to prevent occlusion of the system. Initially, a pocket for inserting the chamber was made with a 3-5-cm long skin incision over the patient’s lower rib cage. The venous catheter was inserted into the subclavian vein using the Seldinger technique with a 12-F peel-away introducer through a subcutaneous tunnel. The peritoneal catheter was inserted into the abdominal cavity with a 16-F peel-away introducer using a similar technique. Then, the PVS system was implanted percutaneously. After checking the position of the entire system by fluoroscopy, the incisions were closed.

Endpoints and statistical analysis

The primary endpoint was improvement of ascites-related symptom. The ascites-related symptom was evaluated with the visual analog scale (VAS) because there is no standard assessment index for this symptom. With the VAS, patients could express their pain level by making a mark on a straight 10-cm line that represents a range from “no pain” (0) on the left side to "unbearable maximum pain” (10) on the right side. Evaluations were conducted twice before enrollment and every other day between days 0 and 28. Independent investigating members of this trial collected the VAS scores from patients. The symptom improvement score was created using a value from subtracting the VAS score from 10. Thus, the scores obtained were graphically plotted, and the areas under the curves (AUCs) for the two groups were compared by analysis of covariance with baseline values as covariates to investigate the superiority of the test group over the control group. The differences in the AUCs for each of the assessments were calculated using a previously used method [22, 23]. The secondary endpoints were improvement in overall QOL, in terms of the EuroQOL-5D (EQ-5D) score (Japanese version) [24] and Short Form-8 (SF-8) score (Japanese version, acute 1-week version) [25], adverse events, and survival time. Improvement in the comprehensive QOL was evaluated in the same way as the primary endpoint by taking the total scores and performing analysis of covariance with baseline values as covariates to investigate the superiority of the test group over the control group. When protocol treatment was terminated within 4 weeks from enrollment, data after termination were not combined for each AUC. Adverse events were evaluated with the CTCAE version 3 up to 4 weeks after enrollment or until protocol treatment was discontinued, and Fisher exact test was used to compare their incidences and grades between the two groups. The Kaplan-Meier method was used to estimate survival, and the log-rank test was used to compare survival between the groups. The level of significance for hypothesis testing was 10% for two-sided tests, and the confidence coefficient was 90%. For the primary endpoint, the difference between the symptom improvement scores for the test group and control group was estimated at approximately 30, with the standard deviation at approximately 40. It was assumed that the total scores would be normally distributed, and with a two-sided significance level of 10%, the size of each group was set at 18 patients to provide 70% power. Therefore, with a 1:1 allocation ratio, the total number of subjects was set to 40.

Enrollment

Forty patients were enrolled between November 2009 and July 2018, and randomly assigned to the test group (n = 20) and control group (n = 20). One patient in the test group and two patients in the control group were excluded from the full analysis set (FAS), because they did not provide a VAS score twice before the enrollment. Thus, the FAS included 37 patients (test group, 19; control group, 18). The primary endpoint and improvement of comprehensive QOL were evaluated using data of the FAS, and all 40 cases were used in the safety analysis set (Fig. 1). There was no significant difference in background characteristics between the groups.

Continuation of protocol treatment

Of 20 patients in the test group, PVS placement was stopped in one patient because of a large amount floating solids in the ascites, the protocol treatment was discontinued because of PVS-related pain in one patient, and the remaining 18 patients died of an underlying disease. As a result, the median duration of protocol treatment in 19 patients of the test group in the FAS set was 56.8 days (range, 3-309 days). The overlapped treatments for 20 patients in the control group were paracentesis (15 cases), diuretic administration (eight cases), albumin preparation administration (five cases), cell-free and concentrated ascites reinfusion therapy (CART) (three cases). In 12 cases, the protocol treatments were discontinued because PVS placements were performed as requested by the patients. Seven patients died due to an underlying disease, and one patient was lost follow-up because of transfer to another hospital. Therefore, the median duration of protocol treatment of 18 patients of the control group in the FAS set was 6.5 days (range, 1-827 days).

Clinical efficacy

Table 1 shows the results of the primary endpoint and EQ-5D and SF-8 as the secondary endpoints. The means and 95% confidence intervals (CIs) of the AUC of the symptom improvement score curves were 133.2 (95% CI: 98.6-167.7) in the test group and 69.9 (95% CI: 34.4-105.4) in the control group, so the primary endpoint was significantly better in the test group than in the control group (P = 0.014).

Concerning the secondary endpoints, the means and 95% CIs of the AUC of the comprehensive QOL evaluated with the SF-8 were 727.3 (95% CI: 614.1-840.6) and 842.6 (90% CI: 670.0-1015.2) in the test group and control group, respectively. There was no significant difference between the groups (P = 0.36). Those evaluated with the EQ-5D were 8.97 (95% CI: 6.50-11.43) and 13.53 (95% CI: 9.78–17.28) in the test group and control group, respectively. Additionally, there was no significant difference between the groups (P = 0.10). Median survival durations were 43.0 days (95% CI: 26.0–61.0 days) in the treatment group and 64.0 days (95% CI: 20.0-253.0 days) in the control group, and there was no significant difference between the groups (P = 0.13).

Adverse events

Twelve patients (test group, eight; control group, four) died within 30 days after enrollment. All deaths were due to disease progression; protocol-treatment-related death was not observed. One hundred twenty-nine adverse events were observed in the test group (grade 1: 42, grade 2: 33, grade 3: 43, grade 4: five, grade 5: six) during the study period, and 68 adverse events were observed in the control group (grade 1: 18, grade 2: 18, grade 3: 28, grade 4: four); there was no significant difference in the frequency of individual adverse events between the groups. Grade 2 or higher adverse events for which a causal relationship with the protocol treatment cannot be ruled out were as follows: grade 3 (anemia, n = 1; pleural effusion, n = 1), grade 2 (diarrhea, n = 1), grade 4 (decrease of the hemoglobin level, n = 2), and grade 3 (increase of the creatinine level, n = 1) in the test group and none in the control group (Table 2).

This randomized, controlled trial was conducted to evaluate the superiority of PVS over the other treatments for refractory ascites in advanced cancer patient. The primary endpoint was the mean AUC of the symptom improvement score curves, and it was significantly larger in the PVS group comparing with the control group, resulting PVS is better treatment to improve the symptom caused by refractory ascites over the other treatments.

The primary organs of malignant tumors in enrolled patients were mainly the pancreas, stomach, colon, lung, and bile duct, and there was no significant difference between the groups. These diseases are considered representative of the diseases causing refractory ascites observed in clinical practice. PVS placements were performed for all patients in the test group, except in one patient in whom PVS placement was stopped because of a large amount of floating solids in the ascites, and PVS treatments were continued for a median of 56.8 days. Thus, there was no issue in the feasibility of PVS placement. In the control group, paracentesis, diuretic administration, albumin preparation administration, and CART were performed individually or in duplicate, and these treatments are considered representative of the treatments used in clinical practice.

The median duration of protocol treatment in the control group was as short as 6.5 days. The reason for this was considered that PVS placement was allowed in this study if the patient in the control group desired it, and 12 of 20 patients received PVS placement during the evaluation period and discontinued the protocol treatment. However, all enrolled patients were informed about PVS and various other treatments, and their desire must be respected. So, this issue was considered unavoidable from an ethical viewpoint.

The primary endpoint, improvement of the ascites-related symptom, was significantly better in the test group than in the control group, and this study met the primary endpoint. This means that PVS not only reduces ascites objectively but it also improves the ascites-related symptom compared to other treatments from the patient’s perspective. Thus, PVS may contribute to the improvement of patients’ QOL.

However, there was no significant difference between the groups in the comprehensive QOL evaluated by the EQ-5D and SF-8. It was thought that PVS cannot improve the comprehensive QOL, although PVS improves symptoms associated with ascites because patients with refractory ascites have various symptoms caused by diseases other than ascites.

Survival was not significantly different between the groups, and the median survival was short. However, the median survival in this study was comparable to that of a retrospective study (JIVROSG-0809) of 133 patients (41 days) [21]. This finding clearly indicates that the prognosis of this patient population is poor.

The numbers of observed adverse events were 129 in the test group and 68 in the control group, and the value was about double in the test group compared with the control group. However, the main cause for this discrepancy was considered the short observation period in the control group. Because 12 of 20 patients in the control group discontinued the protocol treatment and received PVS placement, the median duration of the protocol was only 6.5 days in the control group, which was extremely short compared to 56.8 days in the test group. Additionally, the most of adverse events observed in test group have been previously reported, and they were judged clinically allowable.

Twelve deaths occurred within 30 days after enrollment, but a causal relationship with the protocol treatment was ruled out in all cases, which was considered to reflect the poor prognosis of the patients in this study. Therefore, this study showed that PVS placement significantly improved the ascites-related symptom when compared with other treatments from the patient’s perspective, but it could not improve the patient’s comprehensive QOL.

This study has several limitations. First, VAS was used to evaluate the primary endpoint since there was no established method to evaluate the ascites-related symptom. Second, although this study was a prospective randomized controlled trial, the number of cases was small, level of significance for hypothesis testing was 10% for two-sided tests, and a 90% CI was used for the level of significance. Third, from an ethical viewpoint, performing PVS placement when desired by the patient in the control group might have shortened the protocol duration of the control group and lowered the evaluation value of the primary endpoint and comprehensive QOL.

Despite these limitations, this study evaluated PVS as palliative care in a prospective, randomized controlled trial and demonstrated the efficacy of PVS for improving the ascites-related symptom from a patient’s perspective. Thus, this study provided important suggestions for treatment selection in patients with refractory ascites.

PVS significantly improves the symptom of pain due to refractory ascites compared to other treatments. Although not as effective as improving the comprehensive QOL, PVS may be considered the standard of care for advanced cancer patients with refractory ascites.

Ethical Approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of National Cancer Center Hospital (Date: September 4, 2009/ No. 2009-032). Written informed consent was obtained from all participants.

Competing interests

The authors declare that they have no conflict of interest.

Authors' contributions

All authors contributed to the study’s conception, design, and the protocol. Material preparation, data collection, and analysis were performed by YA, MSo, YT, SS, TA, TT, MSa, and SM. The first draft of the manuscript was written by YA and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This prospective clinical trials were designed, sponsored, and conducted by the Japan Interventional Oncology Study Group (JIVROSG). Fundings from the Japan Agency for Medical Research and Development (AMED) [16ck0106058h0003 to Y.A.]; and the National Cancer Center Research and Development Fund [26-A-27 and 2020-J-3 to M.S.] are gratefully acknowledged.

Acknowledgements

We thank all patients who participated in this study.

Availability of data and materials

All data and materials are available on request.

This prospective clinical trial was designed, sponsored, and conducted by the Japan Interventional Oncology Study Group (JIVROSG). Fundings from the Japan Agency for Medical Research and Development (AMED) [16ck0106058h0003 to Y.A.]; and the National Cancer Center Research and Development Fund [26-A-27 and 2020-J-3 to M.S.] are gratefully acknowledged.

Becker G, Galandi D, Blum HE (2006) Malignant ascites: systematic review and guideline for treatment. Eur J Cancer 42:589–597. https://doi.org/10.1016/j.ejca.2005.11.018
Smith EM, Jayson GC (2003) The current and future management of malignant ascites. Clin Oncol 15:59–72. https://doi.org/10.1053/clon.2002.0135
Mamada Y, Yoshida H, Taniai N, Bandou K, Shimizu T, Kakinuma D, Mizuguchi Y, Ishikawa Y, Akimaru K, Tajiri T (2007) Peritoneovenous shunts for palliation of malignant ascites. J Nippon Med Sch 74:355–358. https://doi.org/10.1272/jnms.74.355
Zanon C, Grosso M, Aprá F, Clara R, Bortolini M, Quaglino F, Cornaglia S, Simone P (2002) Palliative treatment of malignant refractory ascites by positioning of Denver peritoneovenous shunt. Tumori 88:123–127.
Tueche SG, Pector JC (2000) Peritoneovenous shunt in malignant ascites. The Bordet institute experience from 1975–1998. Hepatogastroenterology 47:1322–1324.
Holm A, Halpern NB, Aldrete JS (1989) Peritoneovenous shunt for intractable ascites of hepatic, nephrogenic, and malignant ascites. Am J Surg 158:162–166. https://doi.org/10.1016/0002-9610(89)90368-1
Qazi R, Savlov ED (1982) Peritoneovenous shunt for palliation of malignant ascites. Cancer 49:600–602. https://doi.org/10.1002/1097-0142(19820201)49:3%3C600::aid-cncr2820490333%3E3.0.co;2-t
Adam RA, Adam YG (2004) Malignant ascites: past, present, and future. J Am Coll Surg 198:999–1011. https://doi.org/10.1016/j.jamcollsurg.2004.01.035
Bieligk SC, Calvo BF, Coit DG (2001) Peritoneovenous shunting for gynecologic malignant ascites. Cancer 91(7):1247–1255
Edney JA, Hill A, Armstrong D (1989) Peritoneovenous shunts palliate malignant ascites. Am J Surg 158:598–601.
Faught W, Kirkpatrick JR, Krepart GV, Heywood MS, Lotocki RJ (1995) Peritoneovenous shunt for palliation of gynecologic malignant ascites. J Am Coll Surg 180:472–474.
Gough IR, Balderson GA (1993) Malignant ascites: a comparison of peritoneovenous shunting and nonoperative management. Cancer 71:2377–2382. https://doi.org/10.1002/1097-0142(19930401)71:7%3C2377::aid-cncr2820710732%3E3.0.co;2-h
Ginès P, Arroyo V, Vargas V, Planas R, Casafont F, Panés J, Hoyos M, Rimola A, Morillas R (1991) Paracentesis with intravenous infusion of albumin as compared with peritoneovenous shunting in cirrhosis with refractory ascites. N Engl J Med 325:829–835. https://doi.org/10.1056/nejm199109193251201
Seike M, Maetani I, Sakai Y (2007) Treatment of malignant ascites in patients with advanced cancer: peritoneovenous shunt versus paracentesis. J Gastroenterol Hepatol 22:2126–2166. https://doi.org/10.1111/j.1440-1746.2006.04793.x
Söderlund C (1986) Denver peritoneovenous shunting for malignant or cirrhotic ascites: a prospective consecutive series. Scand J Gastroenterol 21:1161–1172. https://doi.org/10.3109/00365528608996438
Schumacher DL, Saclarides TJ, Staren ED (1994) Peritoneovenous shunts for palliation of the patient with malignant ascites. Ann Surg Oncol 1:378–381. https://doi.org/10.1007/bf02303809
Hussain FF, Meer ZF, Lopez AJ (2004) Peritoneovenous shunt insertion for intractable ascites: a district general hospital experience. Cardiovasc Intervent Radiol 27:325–328. https://doi.org/10.1007/s00270-004-0146-x
Park JS, Won JY, Park SI, Park SJ, Lee DY (2001) Percutaneous peritoneovenous shunt creation for the treatment of benign and malignant refractory ascites. J Vasc Interv Radiol 12:1445–1448. https://doi.org/10.1016/s1051-0443(07)61707-2
Won JY, Choi SY, Ko H, Kim SH, Lee KH, Lee JT, Lee DY (2008) Percutaneous peritoneovenous shunt for treatment of refractory ascites. J Vasc Interv Radiol 19:1717–1722. https://doi.org/10.1016/j.jvir.2008.09.005
Bratby MJ, Hussain FF, Lopez AJ (2007) Radiological insertion and management of peritoneovenous shunt. Cardiovasc Intervent Radiol 30:415–418. https://doi.org/10.1007/s00270-006-0213-6
Sugawara S, Sone M, Arai Y, Sakamoto N, Aramaki T, Sato Y, Inaba Y, Takeuchi Y, Ueno T, Matsueda K, Mariguchi M, Tsushima T (2011) Radiological insertion of Denver peritoneovenous Shunts for malignant refractory ascites: a retrospective multicenter study (JIVROSG-0809). Cardiovasc Intervent Radiol 34:980–988. https://doi.org/10.1007/s00270-010-0057-y
Takeuchi Y, Arai Y, Sone M, Sugawara S, Aramaki T, Sato R, Kichikawa K, Tanaka T, Morishita H, Ito T, Yamakado K, Baba Y, Kobayashi T (2019) Evaluation of stent placement for vena cava syndrome: phase II trial and phase III randomized controlled trial. Support Care Cancer 27:1081–1088. https://doi.org/10.1007/s00520-018-4397-5
Aramaki T, Arai Y, Takeuchi Y, Sone M, Sato R, Bekku E, Moriguchi M (2020) A randomized controlled trial of the efficacy of percutaneous transesophageal gastro-tubing (PTEG) as palliative care for patients with malignant bowel obstruction: the JIVROSG0805 trial. Support Care Cancer 28:2563–2569. https://doi.org/10.1007/s00520-019-05066-8
Tsuchiya A, Ikeda S, Ikegami N, Nishimura S, Sakai I, Fukuda T, Hamashima C, Hisashige A, Tamura M (2002) Estimating an EQ-5D population value set: the case of Japan. Health Econ 11:341–353. https://doi.org/10.1002/hec.673
Fukuhara S, Suzukamo Y (2004) Manual of the SF-8 Japanese version. Institute for Health Outcomes & Process Evaluation Research, Kyoto

Tables 1-2 is available in the Supplementary Files section.

No competing interests reported.

Table1.tiff
Table 1: Results of the primary endpoint, EQ-5D, and SF-8 The primary endpoint was significantly better in the test group than in the control group (P=0.014). There was no significant difference of the secondary endpoints (EQ-5D, SF-8) between the groups.
Table2.tiff
Table 2: Adverse events (number of events) The was no significant difference in the frequency of individual adverse events between the groups.

Download PDF

Version 1

posted

You are reading this latest preprint version

Efficacy of the peritoneovenous shunt as palliative care for cancer patients with refractory ascites: a randomized controlled trial (JIVROSG-0803 study)

Status:

Version 1

Abstract

Figures

Introduction

Patients And Methods

Results

Discussion

Conclusions

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1