Palliative Interventions in Patients With Peritoneal Metastases and Malignant Bowel Obstruction

Abstract

Peritoneal metastases are a common mode of tumor progression for patients with cancers of the GI tract, genitourinary cancers, peritoneal mesothelioma, and occasionally other primary sites. Once peritoneal metastases develop, patient morbidity and mortality are almost invariably a consequence of disease progression in the abdominal cavity. Options for therapeutic intervention include systemic combination chemotherapy and, in selected patients, operative cytoreduction with intraoperative or perioperative regional chemotherapy. Ultimately, the clinical sequelae of uncontrolled tumor progression in the peritoneal cavity from any primary tumor site include abdominal pain, distention, occasional ascites, weight loss, malignant bowel obstruction (MBO) and inanition. Successful palliative management of patients during the late stages of their disease course is clinically challenging. Intravenous hydration, intestinal decompression via a nasogastric or gastrostomy tube, analgesics, and other supportive measures are the norm. Based on their potent antisecretory properties, the use of somatostatin analogs to treat symptoms associated with MBO has been advocated for many years. There have been several small prospective random assignment trials generally showing some benefit to somatostatin analogs compared with anticholinergic agents or corticosteroids in this patient population. In the article accompanying this editorial Mariani et al reports the results of a phase III double-blind placebo controlled study assessing the efficacy of a single dose of intramuscular lanreotide, a long-acting somatostatin receptor analog, in palliating symptoms associated with MBO in hospitalized patients. Patients with peritoneal carcinomatosis who had symptoms of obstruction ( 2 episodes of vomiting per day or the presence of a nasogastric tube) were eligible for the study. The patient cohort was a group suffering from advanced cancer; none were on active therapy for their underlying malignancy, all had diminished performance status, none were considered operative candidates for palliation, and almost two thirds had nasogastric tubes. The primary outcome was improvement in symptoms based on diaries filled out by the patients through day 7 of the study; there were a number of secondary end points and supportive analyses. The authors conclude that lanreotide “has some efficacy and is safe in the symptomatic treatment of patients” with advanced malignant bowel obstruction. This study, testing a palliative intervention in symptomatic and clinically deteriorating patients with MBO from advanced cancer, exemplifies the many and unique challenges that are encountered with palliative care clinical research, in which the intent of therapy is solely to provide symptomatic improvement without any treatment directed against the underlying neoplastic process. There have been a number of publications that have described the unique considerations in designing and conducting prospective palliative care clinical research including identifying measurable, relevant, and generalizable end points; recruitment of study participants; addressing the unique ethical concerns such as minimizing the burden of participation, and study design. It has been noted that timely accrual to trials in palliative care clinical research is challenging and that eligibility criteria must be defined as broadly as possible to facilitate subject enrollment while maintaining as much patient homogeneity as possible so that the results are generalizable. While Mariani et al have done so by including patients with MBO secondary to any tumor histology, the 80-patient accrual in the current study still required a 5-year interval across over 20 hospitals in three countries (Belgium, France, and the Netherlands). Slow recruitment may also be due to inherent reluctance to enroll patients in a clinical study during the end stage of their malignant disease. Some would argue that these patients represent a vulnerable population who may be incapable of giving informed consent or should not be burdened with participating in a clinical research study while in the process of dying. In fact, 11 of the 80 enrolled patients died after enrollment but before the 7-day study end point. In palliative care research, study end points must be completed as soon as possible after enrollment due to the declining condition of the study population. Mariani et al should be lauded in that the study design included a 7-day end point. In addition, the investigators included a standardized, aggressive, and compassionate concomitant treatment regimen as baseline therapy for both groups, including unrestricted analgesic use, to minimize the chances that any study participant in either arm would suffer from uncontrolled symptoms or withdraw from the study prematurely. Despite this, almost 20% of randomized patients did require additional symptomatic interventions not prescribed in the protocol which confounded the outcome analysis. Study end points in palliative care research should be relevant and generalizable to a greater population of similarly afflicted patients. This patient cohort suffered from MBO secondary to a number of different neoplastic processes that are appreciated to have variable biologic behaviors. Over half the study population had ovarian or JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 30 NUMBER 35 DECEMBER 1