Abstract 3311

Background

Hemorrhage from disease processes or their therapy commonly complicates the course of patients with malignancies. Recombinant activated factor VII (rFVIIa) was developed for prevention or treatment of bleeding in hemophilia patients who have factor inhibitors, but may also have a role in enhancing coagulation and augmenting conventional therapies in refractory bleeding. However, outcomes are not defined for this indication, and randomized studies are challenging. We explored the safety and efficacy of rFVIIa in the treatment of hemorrhage associated with malignancy or stem cell transplantation (SCT) using prospective clinical registry data.

Methods

The Australia and New Zealand Haemostasis Registry (ANZHR) documented all rFVIIa use in non-hemophilia patients in both countries. More than 90 institutions participated in its operation from 2000 to 2009, capturing an estimated 85% of use in these countries. Participating institutions obtained approval from local institutional review boards to provide de-identified data. They were required to report all usage to limit bias and registry data were audited regularly. We identified episodes of rFVIIa use where the primary diagnosis at the time of hemorrhage was a malignancy or SCT. Response to rFVIIa reported by the treating clinician, transfusion requirements, adverse events and 28 day survival were analyzed. Logistic regression and multiple linear regression models were used to identify factors associated with response.

Results

Of 3446 episodes reported to ANZHR, 362 eligible episodes were identified in 325 patients (10.5% the total). Median patient age was 56 years (IQR 39–67); 205 (63%) were male. Episodes were reported from 51 institutions (57% of participating), median 5 (IQR 3–8) per institution. Primary diagnosis was hematological in 173 cases (48%); non-hematological malignancy in 189 (52%). 321 (88.6%) were receiving disease-directed therapy (initial therapy in 200 [55% of the total], 19 [5%] supportive care, unreported in 22 [6%]). Intent of care was curative in 218 (60%), palliative in 122 (34%), 22 (6%) unreported. 33 (9.1%) were undergoing allogeneic SCT and 5 (1.4%) autologous SCT. Primary sites of bleeding were gastrointestinal in 165 (46%), pulmonary 47 (13%), intracranial 36 (10%), genitourinary 34 (9%), hepatic 28 (8%), others 52 (14%). Causes of bleeding were 116 (34%) surgical (including resection of tumor or metastases), disease-induced coagulopathy 48 (14%), disease invasion 62 (18%), mucositis (including graft-versus-host disease) 53 (15%), others 65 (19%); 6 (2%) received rFVIIa prophylactically prior to procedures. In 255 episodes (70%) a single dose of rFVIIa only was administered. Median dose was 90μg/kg (IQR 77–100). Prior to administration median (IQR) hemoglobin was 82 g/L (69–93); platelet count 82 × 109/L (43–127), fibrinogen 2.1 g/L (1.4–3.1), INR 1.4 (1.2–1.8), pH 7.3 (7.2–7.4), temperature 36.7°C (36.0–37.1), and patients had received 5 (2–12) units red cells, 2 (0–4) platelet doses, 4 (0–8) plasma units, and 0 (0–8) cryoprecipitate doses. Bleeding stopped or decreased in 175 (60%) of cases where response was reported (prophylactic cases were excluded from response assessment). Transfusion requirements fell to 2 (0–4) red cells, 0 (0–2) platelets, 0 (0–3) plasma, and 0 (0–8) cryoprecipitate following the first dose. Response decreased with low pH in univariate analysis (p=0.014). In multivariate analysis, low pH and hematological malignancy were independent predictors of poor response (p=0.001). 18 patients (5.5%) had thromboembolic events reported within 28 days of administration. 202 patients (57%) survived to 28 days. Survival was significantly higher in rFVIIa responders (71% vs 38% p<0.001, Figure 1).

Conclusions

This is the first comprehensive exploration of rFVIIa in hemorrhage associated with malignancy. Usage in this setting was higher than anticipated. Reported efficacy and safety were promising in this heterogeneous group with severe refractory bleeding and elevated baseline risk of thrombosis. Clinician-reported bleeding response was supported by reduced transfusion requirements and survival benefit. Response was associated with pH and type of malignancy, suggesting that these factors may assist in patient selection. Use of rFVIIa may be an effective adjunct to conventional therapies in carefully selected cancer patients with refractory hemorrhage.

Disclosures:

Off Label Use: Recombinant Activated Factor VII, for bleeding associated with malignancy.

This icon denotes a clinically relevant abstract

Author notes

*

Asterisk with author names denotes non-ASH members.