Safety and efficacy of trabectedin when administered in the inpatient versus outpatient setting: Clinical considerations for outpatient administration of trabectedin

Background

The results of the randomized, phase 3 ET743‐SAR‐3007 trial demonstrated that trabectedin had a significantly longer progression‐free survival (PFS) compared with dacarbazine in patients with advanced leiomyosarcoma/liposarcoma after the failure of prior chemotherapy. Patients randomized to trabectedin received a 24‐hour intravenous infusion either in an inpatient or outpatient setting. Herein, the authors reported the safety, efficacy, and patient‐reported outcomes based on first infusion site of care.

Methods

Patients were randomized 2:1 to trabectedin (at a dose of 1.5 mg/m²) or dacarbazine (1 g/m² over 20‐120 minutes) with overall survival (OS) as the primary endpoint and PFS, time to disease progression, objective response rate, duration of response, safety, and patient‐reported symptom scoring as secondary endpoints. The setting of the trabectedin infusion was based on institutional preference and categorized based on the setting of the first infusion.

Results

Of the 378 patients who were treated with trabectedin, 100 (27%) and 277 (73%), respectively, first received trabectedin in the inpatient and outpatient setting. No differences were observed with regard to PFS or OS based on site of care. The median PFS was 4.1 months versus 4.2 months (hazard ratio, 0.90; P = .49) for inpatients versus outpatients, respectively, and the median OS was 14.3 months versus 13.7 months (hazard ratio, 0.89; P = .40), respectively. Grade 3/4 adverse events (classified according to the National Cancer Institute Common Terminology Criteria for Adverse Events [version 4.0]) were reported in 87 inpatients (87%) compared with 219 outpatients (79%); grade 3/4 serious adverse events were reported in 43 inpatients (43%) and 92 outpatients (33%). Extravasation occurred in 0 inpatients and 5 outpatients (2%), whereas the incidence of catheter‐related complications was similar between groups (16% vs 15%).

Conclusions

Although the majority of patients who were randomized to trabectedin received outpatient therapy, the outcomes of the current study suggested equivalent safety and efficacy in either setting.

The majority of patients randomized to trabectedin in the ET743‐SAR‐3007 trial have received the drug in the outpatient setting. The results of the current analysis suggest that trabectedin has equivalent efficacy and comparable safety when administered in an inpatient or outpatient setting.

Development and usability of a new subcutaneous auto-injector device to administer hydroxyprogesterone caproate to reduce the risk of recurrent preterm birth

Background

Current administration of hydroxyprogesterone caproate (HPC) by intramuscular injection is associated with limitations, including the potential for human error and contamination, patient anxiety, and increased risk of needlestick injury.

Objective

To describe the design of an auto-injector for subcutaneous (SC) administration of HPC and the results of studies that evaluated the target user's understanding of the proper use of this device.

Materials and methods

A single-use, prefilled, fixed-dose, disposable auto-injector intended for the SC administration of HPC was developed, and its usability by health care providers was evaluated in 3 formative (N=32, 64 injections) and 3 validation studies (N=45, 90 injections). These studies consisted of one-on-one testing sessions performed in a simulated home environment. Analyses were based on observed use error or use difficulty during the performance of specific tasks, including those considered critical (associated with high severity harms).

Results

In the formative studies, the majority of participants correctly administered an injection with the auto-injector, but prior training improved performance. Specific errors were noted, including holding the device at the injection site for a period inconsistent with its instructions for use (IFU). The IFU was modified to reduce potential occurrence of these errors. Use errors were subsequently observed on critical tasks in the first and second validation studies, including hold-time errors that were attributed to using visual cues rather than counting seconds. For the third validation study, the IFU was modified to focus on visual cues and all users were able to successfully perform the injection per the IFU.

Conclusion

An auto-injector device for SC administration of HPC for reduction in risk of recurrent preterm birth was successfully developed through iterative design and validation testing. The device design provides high usability and acceptance of this device by health care professionals.

A cell-based artificial antigen-presenting cell coated with anti-CD3 and CD28 antibodies enables rapid expansion and long-term growth of CD4 T lymphocytes

We compared the ability of two genetically modified myeloid cells, K562 and U937, to serve as artificial antigen-presenting cells (aAPC). Both aAPC were stably transfected with the low-affinity Fcgamma receptor CD32 (K32/U32 cells). K32 cells loaded with anti-CD3 and anti-CD28 Ab (K32/CD3/28) induced more rapid CD4 T-cell expansion than CD3/28-coated beads. In contrast, U32/CD3/28 induced high levels of CD4 T-cell thymidine uptake but were unable to sustain long-term T-cell expansion. K32 cells, but not U32 cells, loaded with anti-CD3 alone also stimulated CD4 T-cell growth and IL-2 secretion, indicating the expression of additional costimulatory molecules on K32 cells. We found constitutive expression of B7-H3 and a strong upregulation of mRNA encoding for IL-15, PD-L1, and PD-L2 after coculture with CD4 T cells activated by K32/CD3/28 but not U32/CD3/28. We conclude that K32 aAPCs are a robust system for clinical scale ex vivo expansion of CD4 T cells.

Development of oral vaccines to stimulate mucosal and systemic immunity: barriers and novel strategies

Many questions regarding the induction of mucosal and humoral immunity through oral vaccination exist. Efficacy is dependent on the physicochemical properties of the antigen, the gastrointestinal environment, the presence of adjuvants, and the mode of delivery. Understanding how these factors interrelate will be critical to the development of new oral vaccines. A number of approaches are currently being studied to enhance the immune response. These include chemical conjugation, immunization with recombinant bacteria and viruses, and mucosal adjuvants. Vaccine delivery systems prepared from natural or synthetic polymers is a particularly promising area because many of the current methods to induce mucosal stimulation can be incorporated within these systems. Thus, the polymeric delivery system functions as a platform to facilitate uptake by M-cells and prolong antigen presentation and stimulation of the Peyer's patches. This Review examines some of the physiological and immunological barriers associated with oral vaccination and discusses novel strategies to overcome such barriers.

Evaluation of an orally administered vaccine, using hydrogels containing bacterial exotoxins of Pasteurella haemolytica, in cattle

Poly(methacrylic acid) hydrogels were tested for oral delivery of a vaccine against Pasteurella haemolytica infection in cattle. Culture supernatants of P haemolytica, the most common bacterium associated with pneumonia in cattle, were used as the antigens in the vaccine. Hydrogels containing culture supernatants were administered orally to calves. Calves were then challenge-exposed with virulent P haemolytica. Calves were euthanatized 3 days after challenge exposure. The lungs of each calf were scored for severity and size of pneumonic lesions. Results indicated that vaccinated calves had smaller, less severe pneumonic lesions and lived longer than nonvaccinated calves. These results indicated that hydrogels can be used to deliver vaccines orally to calves to enhance resistance to pneumonia caused by P haemolytica.

Use of ultrasound imaging and fluoroscopic imaging to study gastric retention of enzyme-digestible hydrogels

Ultrasound and fluoroscopic imaging techniques were used to monitor the gastric retention of enzyme-digestible hydrogels in the canine stomach. When water was present in the stomach, ultrasound imaging was very effective in monitoring the position of the hydrogel in the stomach, solvent penetration into the gel, and the gastric tissue-gel interactions during peristalsis. Rubbery or fully swollen hydrogels appeared as sonolucent objects with ultrasound imaging. Partially swollen hydrogels displayed a sonolucent outer layer due to solvent penetration and a centrally located bright echo resulting from the acoustic impedance mismatch at the glassy/rubbery interface. The degree of gastric tissue-gel interactions during peristalsis was inversely related to the extent of lumenal distention with water. The effectiveness of peristaltic contractions in driving the hydrogel toward the pyloric sphincter increased as the water was emptied from the stomach. In the absence of water, imaging of the gel with ultrasound became difficult. For this reason, gels were loaded with diatrizoate meglumine/sodium diatrizoate to visualize in real-time using fluoroscopic imaging. Fluoroscopic imaging allowed only indirect assessment of the hydrogel movement during peristalsis and the degree of hydrogel swelling. The gastric retention of the hydrogel under fasted conditions was influenced by the degree of gel deformation in response to peristaltic contractions. Hydrogels with a low degree of deformation during peristalsis showed long gastric retention times. The utilization of ultrasound imaging and fluoroscopic imaging for monitoring dynamic events in the stomach provided information on hydrogel properties which are important to gastric retention. The use of these imaging techniques in the development of long-term oral drug delivery systems is described.

Biochemical and mechanical characterization of enzyme-digestible hydrogels

Albumin-cross-linked hydrogels were prepared by free radical polymerization using 1-vinyl-2-pyrrolidinone as a monomer and functionalized albumin as a crosslinking agent. The degree of chemical cross-linking was controlled by varying the degree of albumin functionality and the concentration of albumin. With emphasis placed on the potential use of these hydrogels for long-term oral drug delivery, gel characterization studies examined both the swelling and the mechanical properties in the absence and presence of pepsin. In the absence of pepsin, the equilibrium swelling ratio in simulated gastric fluid ranged from 17 to 55, depending on the degree of albumin functionality and the albumin concentration. Swelling was pH dependent at pH's greater than 7. The uptake of solvent into the dried hydrogels was determined to be Fickian. The integrity of swelling gels was dependent on the concentration of the functionalized albumin as well as on the degree of albumin functionality. In the presence of pepsin, a predominance of either surface or bulk degradation was observed, depending on the functionality of the albumin used as a cross-linker. Gel integrity during pepsin degradation also showed a marked dependence on the albumin functionality.