Title: Intravenous Bevacizumab Reduces Transfusion Requirements and Endoscopic Interventions in Patients With Gastric Antral Vascular Ectasia and Small Bowel Angioectasia
Abstract: Pharmacologic options for management of gastrointestinal (GI) bleeding from small bowel angioectasias (SBA) and gastric antral vascular ectasia (GAVE) are limited.1Jackson C.S. Strong R. Gastrointestinal Angiodysplasia: Diagnosis and management.Gastrointest Endosc Clin N Am. 2017; 27: 51-62Abstract Full Text Full Text PDF PubMed Scopus (33) Google Scholar, 2Fuccio L. Mussetto A. Laterza L. et al.Diagnosis and management of gastric antral vascular ectasia.World J Gastrointest Endosc. 2013; 5: 6-13Crossref PubMed Google Scholar, 3Junquera F. Saperas E. Videla S. et al.Long-term efficacy of octreotide in the prevention of recurrent bleeding from gastrointestinal angiodysplasia.Am J Gastroenterol. 2007; 102: 254-260Crossref PubMed Scopus (99) Google Scholar, 4Junquera F. Feu F. Papo M. et al.A multicenter, randomized, clinical trial of hormonal therapy in the prevention of rebleeding from gastrointestinal angiodysplasia.Gastroenterology. 2001; 121: 1073-1079Abstract Full Text Full Text PDF PubMed Scopus (211) Google Scholar, 5Ge Z.Z. Chen H.M. Gao Y.J. et al.Efficacy of thalidomide for refractory gastrointestinal bleeding from vascular malformation.Gastroenterology. 2011; 141 (e1–4): 1629-1637Abstract Full Text Full Text PDF PubMed Scopus (175) Google Scholar We have previously demonstrated the efficacy of intravenous (IV) bevacizumab in controlling refractory GI bleeding in patients with hereditary hemorrhagic telangiectasia (HHT).6Iyer V.N. Apala D.R. Pannu B.S. et al.Intravenous bevacizumab for refractory hereditary hemorrhagic telangiectasia-related epistaxis and gastrointestinal bleeding.Mayo Clin Proc. 2018; 93: 155-166Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar Given the morphologic similarities in GI angioectasias between GAVE/SBA and HHT, we hypothesized that IV bevacizumab would result in similar improvements in GAVE/SBA. This novel study evaluates the role of IV bevacizumab for treatment of GI bleeding secondary to GAVE/SBA. All patients with GAVE/SBA treated with IV bevacizumab at the Mayo Clinic in Rochester, Minnesota, from September 1, 2015, to June 30, 2019, were included. Further details about the study methodology, inclusion/exclusion criteria, dosing protocol (Supplementary Figure 1), and statistical analysis are provided in the Supplementary Material. Twenty-one patients with at least 6 months of follow-up (after bevacizumab initiation) were identified (Table 1), 13 (61.9%) of whom required at least 1 top-up of IV bevacizumab because of recurrent GI bleeding and worsening anemia (Supplementary Figure 2).Table 1Demographic and Baseline Characteristics at the Time of Initiation of IV Bevacizumabn (%)Age, median (IQR, range)69 (62–77, range 39–85)Female sex9 (42.9)Comorbidities Hypertension9 (42.9) Diabetes mellitus type 211 (52.4) Congestive heart failure5 (23.8) Liver cirrhosis9 (42.9) Hypothyroidism3 (14.3) Aortic stenosis/mechanical aortic valve7 (33.3) Scleroderma5 (23.8)Smoking history12 (57.1) Active smokers at the time of initiation of therapy3 (14.3) Pack-years30 (IQR 26.25–40)Primary source of blood loss GAVE8 (38.1) Small bowel angioectasia6 (28.6) Both7 (33.3)Hemoglobin (IQR) (g/dL)8.3 (7.6–8.9)Serum ferritin (IQR) (μg/L)51 (16–102)Serum iron (IQR) (μg/L)32 (24–80)Oral anticoagulant3 (14.3)Antiplatelet therapy Single agent (aspirin)2 (9.5) DAPT (aspirin and clopidogrel)1 (4.8)Combined anticoagulant and antiplatelet therapy0 (0)Medan duration of follow-up20.7 (IQR 14.7–25.6; range 4.3–37.5)DAPT, dual antiplatelet therapy. Open table in a new tab DAPT, dual antiplatelet therapy. The median number of red blood cell (RBC) units decreased from 13 (interquartile range [IQR] 7–25; range 5–43) to 0 (IQR 0–0; range 0–10, P < .0001) 6 months before and after treatment, and from 20 (IQR 12–50; range 5–75) to 2 (IQR 0–6.3; range 0–14, P < .0001) 12 months before and after treatment. A total of 11 (52.4%) remained transfusion-free at 6 months and 3 until the date of last follow-up. Overall, 90% (n = 19) and 86% (n = 18) achieved a positive treatment response (≥50% reduction in RBC transfusion needs) at 6 and 12 months of follow-up, respectively. The median amount of IV iron infusion decreased from 1020 (IQR 510–2335; range 0–10815) mg to 0 (IQR 0–1020; range 0–4590) mg (P = .0122) 12 months before and after therapy. The median number of endoscopic procedures decreased from 2 (IQR 1–3.5; range 0–8) to 0 (IQR 0–1; range 0–3) (P = .0029) during the 6 months before and after treatment and from 5 (IQR 3–8; range 0–13) to 0 (IQR 0–1; range 0–4) (P < .0001) during the 12 months before and after treatment. Nine (64.3%) of 14 patients with at least 12 months of follow-up did not require any further endoscopic interventions in the 12 months following initiation of therapy. The median hemoglobin concentration increased from 8.3 (IQR 7.6–8.9) g/dL to 10.4 (IQR 10.1–11.2) (P < .0001) and 9.8 (IQR 8.5–11) (P = .0022) at 6 and 12 months of follow-up, respectively. The difference between iron and ferritin levels before and after treatment was not statistically significant. One patient had worsening of known hypertension, which was controlled with clonidine. Another patient had significant epistaxis requiring transfusion of 2 RBC units; however, this was ultimately controlled with endoscopic laser cauterization. Following this, bevacizumab was resumed with no further adverse events. A total of 10 patients died during follow-up after a median of 9.5 (IQR 4.0–17.6; range 0.2–25.2) months; 2 of whom were not included in the final cohort of 21 patients, as they did not complete 6 months of follow-up. In no case was the cause of death directly linked to bevacizumab treatment (Supplementary Table 1). We present the first series of patients treated with IV bevacizumab for refractory GI bleeding secondary to GAVE/SBA. We found significant reduction in the need for endoscopic procedures and a 10-fold reduction in RBC transfusion requirements 1 year posttreatment with more than half patients remaining transfusion-free at 6 months. This dramatic improvement is even more remarkable when one considers that IV bevacizumab was used as a therapy of last resort in this cohort of patients with refractory GI bleeding and transfusion-dependent anemia. It is worth noting that requirements for "top-up" or maintenance bevacizumab dosing appeared to vary considerably among patients after completion of the initial dosing regimen (Supplementary Figure 2). These findings mirror our previous experience with maintenance dosing in HHT-related bleeding.6Iyer V.N. Apala D.R. Pannu B.S. et al.Intravenous bevacizumab for refractory hereditary hemorrhagic telangiectasia-related epistaxis and gastrointestinal bleeding.Mayo Clin Proc. 2018; 93: 155-166Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar The dosing strategy for initial as well as maintenance doses was adopted from our experience with patients with HHT, but may need to be refined to address the needs of this unique patient population with GAVE/SBA. The present study has several limitations. The first limitation is the retrospective, nonrandomized and nonblinded nature of the study. Despite this, patients received bevacizumab using a standardized dosing protocol by a single provider (VNI) with considerable experience in its use among patients with HHT with GI bleeding. Patients also underwent careful follow-up at regular intervals with serial laboratory assessment, which helped limit the shortcomings of an otherwise retrospective study. Another limitation is that baseline and follow-up hemoglobin measurements can be influenced by blood transfusions and iron infusions; however, we were able to show a marked reduction in RBC transfusion as well as IV iron requirements on follow-up. These "hard" endpoints are less likely to be influenced by short-term fluctuations in hemoglobin values due to RBC transfusions or IV iron infusions. Strengths of our study include the relatively large number of patients with refractory GI bleeding and transfusion dependence, the standardized dosing protocol that was used, as well as the median follow-up duration approximating 2 years. These attributes of the study give us a high level of confidence about the sustained beneficial effects of bevacizumab on bleeding reduction. In conclusion, we report the first study on the use of IV bevacizumab as a novel treatment for the management of refractory GI bleeding and transfusion-dependent anemia from GAVE/SBA. IV bevacizumab treatment was associated with a marked reduction in the rate of RBC transfusions, IV iron infusions, and endoscopic interventions throughout the duration of follow-up. We acknowledge Sahil Khanna, MBBS, Louis M. Wong Kee Song, MD, and Shounak Majumder, MD, of the Mayo Clinic Division of Gastroenterology and Hepatology for their valuable insights and contributions to the successful completion of this project. Author Contributions: Conception and design: HA, YA, PK, ER, VI. Analysis and interpretation HA, YA, PK, ER, SK, LW, SM, KP, VI. Drafting the manuscript: HA, YA, PK, ER, SK, LW, SM, KP, VI This retrospective study was approved by the institutional review board. Refractory GAVE/SBA related GI bleeding was diagnosed if patients remained persistently anemic despite at least 3 esophagogastroduodenoscopic/colonoscopic procedures, or double-balloon endoscopic therapy along with ongoing medical therapy (such as reduction/cessation of antithrombotic therapy) that failed to control GI bleeding. We included all patients with refractory GAVE/SBA receiving IV bevacizumab who had at least 6 months of follow-up after therapy initiation. Absolute contraindications for bevacizumab included ischemic colitis, peptic ulcer disease, thrombocytopenia, recent systemic chemotherapy (within 6 months), recent intracranial or internal bleeding (within 1 year), nonhealing wounds, active infection, decompensated congestive heart failure, and major surgery (within 3 months). Relative contraindications included uncontrolled hypertension and venous thromboembolism (within 12 months). Patients on antithrombotic therapy and patients with cirrhosis were not excluded. The initial dosing cycle consisted of 4 to 6 doses (5 mg/kg each) administered IV once every 2 weeks followed by 4 additional doses given IV on a monthly basis. Thus, each patient received between 8 and 10 doses of IV bevacizumab over a roughly 5- to 6-month initial treatment period. Based on our experience with patients with HHT, it was anticipated that most patients in this cohort would require maintenance or "top-up" bevacizumab dosing at some time point due to recurrent bleeding. The decision to give further maintenance bevacizumab doses was individualized. Given the natural variability in serial hemoglobin values, we used a ≥1.5 g/dL drop in hemoglobin (from the patient's own peak hemoglobin achieved post bevacizumab) to initiate "top-up" dosing. This approach allowed patients to receive additional bevacizumab before they became significantly anemic and at the same time avoided unnecessary treatment for minor hemoglobin fluctuations in patients without obvious signs of rebleeding. Iron deficiency was managed aggressively in all patients using a combination of oral and/or IV iron. This was done both during initial bevacizumab dosing as well as on follow-up, with IV ferumoxytol (FERAHEME) used as the preferred agent. Monthly laboratory testing, including hemoglobin, iron, and ferritin levels, were performed, and patients were instructed to maintain close follow-up so that "top-up" bevacizumab doses could be administered before significant deteriorations in hemoglobin and/or ferritin/iron levels occurred. Each "top-up" maintenance dosing cycle consisted of 1 to 2 infusions (5 mg/kg IV) 2 weeks apart. In some cases, a higher bevacizumab dose (7.5 mg/kg) was used for 1 to 2 doses if a suboptimal response was noted with the previous 5 mg/kg top-up cycle. We defined a positive treatment response as a ≥50% reduction in RBC transfusion needs in the 6 months following initiation of IV bevacizumab compared with the 6 months preceding therapy initiation. The data were collected and analyzed using JMP statistical software. Baseline characteristics were described using counts and percentages for categorical variables and median (IQR; 25th–75th, range) for nonparametric continuous variables. Characteristics and secondary outcomes were assessed using χ2 for categorical variables, and Kolmogorov-Smirnov test for nonparametric variables. A P value of ≤ .05 was considered statistically significant. Details regarding the treatment regimen are provided in the supplementary file .Supplementary Figure 2Maintenance bevacizumab dosing and follow-up data.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Table 1Mortality in Patients Treated with IV Bevacizumab during the Study PeriodPatientAgeSexNumber of bevacizumab doses receivedResponse to bevacizumabPertinent comorbidities in addition to GAVE and SBACause of deathTime to death from last bevacizumab dose177M9YesRecurrent pulmonary embolism with CTEPH, bladder cancer, APLS, AIHADied after being on hospice for 2 months; unknown cause of death; no autopsy available8 months252M22YesEnd-stage ILD secondary to scleroderma, ESRD, systemic sclerosis, multivessel CAD on antiplatelet therapyCardiac arrest25 months (12 from last dose)364F5YesSystemic sclerosis, PAH, esophageal dysmotility with recurrent aspiration pneumonia, MGUS, atrial fibrillationAspiration pneumonia3 months469F24YesLiver cirrhosis, scleroderma, insulin-dependent diabetes mellitusAcute hypoxemic respiratory failure18 months (13 from last dose)585M8YesStroke, atrial fibrillation on anticoagulation, MGUS, CKD 4, recurrent aspiration pneumoniaUnknown, no outside records9 months (7 from last dose)683F7YesCKD 4, insulin-dependent diabetes mellitus, hypertension, congestive heart failure, mechanical aortic valve on warfarinUnknown, no outside records7 months (4 from last dose)769M7YesTransformed acute myeloid leukemia from myelodysplastic syndrome, atrial fibrillationAcute myeloid leukemia11 months (5 from last dose)883F6YesEnd-stage ischemic cardiomyopathy, pulmonary hypertension, atrial fibrillation, torrential tricuspid regurgitationAcute decompensated heart failure4 months (1 from last dose)970M1Unknown due to short follow-up periodCKD, ischemic cardiomyopathy, CKD 4, severe mitral regurgitation, severe aortic stenosisAcute decompensated heart failure6 days following initial dose1058F16No, bevacizumab was discontinued 13 months before death due to lack of responseESRD, non-Hodgkin's lymphoma of the appendix on chemotherapy, insulin-dependent diabetes mellitusCardiac arrest secondary to end-stage renal disease without dialysis18 months (13 from last dose)NOTE. Two of the patients included in this table (3 & 9) died during the initial bevacizumab dosing cycle. They were excluded from data analysis in this study, as they did not complete the initial bevacizumab dosing cycle and had less than 6 months of follow-up.AIHA, autoimmune hemolytic anemia; APLS, antiphospholipid syndrome; CAD, coronary artery disease; CKD, chronic kidney disease; CTEPH, chronic thromboembolic pulmonary hypertension; ESRD, end-stage renal disease; ILD, interstitial lung disease; MGUS, monoclonal gammopathy of undetermined significance; PAH, pulmonary arterial hypertension. Open table in a new tab NOTE. Two of the patients included in this table (3 & 9) died during the initial bevacizumab dosing cycle. They were excluded from data analysis in this study, as they did not complete the initial bevacizumab dosing cycle and had less than 6 months of follow-up. AIHA, autoimmune hemolytic anemia; APLS, antiphospholipid syndrome; CAD, coronary artery disease; CKD, chronic kidney disease; CTEPH, chronic thromboembolic pulmonary hypertension; ESRD, end-stage renal disease; ILD, interstitial lung disease; MGUS, monoclonal gammopathy of undetermined significance; PAH, pulmonary arterial hypertension.
Publication Year: 2020
Publication Date: 2020-03-01
Language: en
Type: article
Indexed In: ['crossref', 'pubmed']
Access and Citation
Cited By Count: 29
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot