A Phase 2 Study of Methylene Blue Photodynamic Therapy for Treatment of Breast Abscesses

Breast abscesses consist of a localized, walled-off collection of infected fluid within the breast, with symptoms including breast pain and inflammation, as well as nipple discharge, fever, chills, and nausea. If left untreated, breast abscesses can result in serious complications, including fistula formation and sepsis. The standard of care therefore involves draining the purulent fluid, generally under ultrasound guidance, and systemic antibiotic administration. Although this drainage procedure is typically technically successful, reported recurrence rates vary widely. These recurrent breast abscesses can require repeated drainage procedures, surgical intervention, extended hospital stays, and aggressive antibiotic regimens.

Photodynamic therapy (PDT), which relies upon the combination of photosensitive dyes known as photosensitizers, excitation by visible light, and molecular oxygen to generate bactericidal reactive oxygen species, represents a potentially powerful adjunct to standard of care drainage. The investigators have successfully completed a Phase 1 clinical trial assessing the safety and feasibility of methylene blue (MB) PDT at the time of percutaneous intra-abdominal abscess drainage. No adverse or serious adverse events were observed, and the procedure was well tolerated by all subjects. Based on the success of this Phase 1 clinical trial, the current study aims to assess efficacy of MB-PDT at the time of breast abscess drainage.

The investigators propose a randomized, controlled Phase 2 clinical trial, which will be conducted in accordance with the principles of good clinical practice and with approval by both the FDA and local IRB. Patients who have been diagnosed with a drainable abscess, that meet all inclusion/exclusion criteria, and have the approval of their primary care team, will be offered enrollment in the study at initial presentation. Consented subjects will be prospectively randomized using block randomization to one of two arms: (1) standard of care (SOC) percutaneous drainage and antibiotics alone or (2) PDT + SOC. The primary outcome is time to resolution of clinical symptoms. The secondary outcome is imaging evidence of abscess resolution, based on ultrasound tomography. Tertiary outcomes include abscess recurrence, need for repeated intervention, and cure rate across treatment groups.

The study population will consist of 50 subjects, allocated equally between the two groups. Sample size calculations are based on a hypothesized reduction in the time to resolution of clinical symptoms. The investigators use retrospective data from patients that received SOC at our institution over the course of 15 years to represent SOC, and the 18 subjects treated in our Phase 1 clinical trial in abdominopelvic abscesses to represent PDT. Retrospective standard of care patients had median time to resolution of 16 days (95% CI 14-18 days), while subjects treated with MB-PDT had a median time to resolution of 7 days (95% CI 3-14 days). This corresponds to a hazard ratio for resolution of symptoms of 3.5. The investigators target 90% power at a more conservative hazard ratio of 3.0 to determine the sample size in each arm. A sample size of 25 subjects per arm provides 90% power to detect a hazard ratio of 3.0 at a two-sided alpha=0.05 level of significance. With equal allocation between groups, this results in a total sample size of 50 participants, given that the investigators assume to fully observe the primary outcome in all subjects.

All subjects will receive standard of care ultrasound-guided percutaneous breast abscess drainage. The site will be prepared and dressed following standard of care sterile procedure, and a needle will be advanced under ultrasound guidance into the abscess. The abscess will be drained by syringe aspiration, and the cavity will be flushed with sterile saline. Aspirated abscess fluid will be delivered to the Clinical Microbiology Laboratory for processing.

For subjects randomized to the PDT arm, sterile methylene blue (MB) will then be administered to the abscess cavity using the same needle/catheter utilized for standard of care drainage. After a 10-minute incubation interval, MB will be aspirated and the cavity flushed with sterile saline. The abscess cavity will then be filled with sterile 0.1% lipid emulsion to gently distend the cavity, and through light scattering, homogenize the light dose to the walls of the cavity. A sterile optical fiber will be advanced to the approximate center of the abscess cavity via the same catheter/needle used for standard of care under image guidance. The proximal end of the fiber will be coupled to the output of a clinical laser system emitting light at 665 nm. Laser power will be delivered to the cavity at a constant fluence rate of 20 mW/cm^2. The laser power required to obtain the desired fluence rate at the wall will be calculated based on the abscess size. Illumination will be delivered for ~20 minutes, resulting in a total delivered fluence of 25 J/cm^2.

Immediately after laser irradiation, the Intralipid solution will be aspirated and the cavity again flushed with sterile saline. SOC drainage will then be completed, including cleaning of the skin surface and wound dressing.

All subjects will receive ultrasound tomography of their affected breast using an FDA-approved ultrasound tomography system located at our institution. This will be performed immediately pre- and post-intervention (SOC or PDT+SOC), and at follow-up appointments. From these images, the investigators will first extract the abscess volume and viscosity of the infected fluid on pre-procedure imaging. At each timepoint, abscess volume, abscess wall stiffness and density, and surrounding breast tissue stiffness and density will be extracted from imaging data.

A Cox proportional hazards model will be used to model time to clinical symptom resolution. A 1-degree of freedom likelihood ratio test comparing time to resolution between the control group (SOC alone) and the PDT group (PDT+SOC) will be used for the primary analysis. Abscess recurrence and need for repeated intervention will be compared between treatment groups using a Cox proportional hazards model. Cure rate will be compared between groups using Fisher's exact test. Imaging outcomes (breast abscess volume, density, stiffness) will be compared between groups using linear mixed-effects models, and will further be compared to radiologist measurements using Wilcoxon matched-pairs signed rank tests. Multivariable Cox proportional hazards models will be used to examine the effects of anatomical features, risk factors, and subject demographics on long-term treatment responses.

If successful, this research could demonstrate efficacy of PDT as an adjunct treatment for breast abscess patients that significantly reduces time to symptom resolution. The investigators hypothesize that this could also lead to reduced abscess recurrence. Ultimately, this would promote early recovery, reduce repeated intervention, and lead to lower overall health care costs for patients undergoing ultrasound-guided percutaneous breast abscess drainage.