Adipose-derived stromal vascular fraction injection for radiation-induced bladder fibrosis: a first-in-human pilot study

Abstract

Pelvic radiotherapy can lead to loss of bladder compliance, detrusor overactivity, and superficial vascular proliferation. Animal studies have demonstrated a reduction in radiation-induced inflammation and fibrosis following administration of mesenchymal stem cell products. We performed a first-in-human pilot study to assess the safety and feasibility of intravesical adipose stromal vascular fraction injection in an 82-year-old man with a history of brachytherapy for prostate cancer. Following FDA and IRB approval, subcutaneous adipose tissue harvested under local anesthesia was immediately processed using the SVF-2 device (GID Bio). The cell product, consisting of 4.58 × 10⁷ stromal vascular fraction cells resuspended in 20 cc/s of lactated ringers, was injected into the bladder wall using a flexible cystoscope. Urinary symptoms, urologic flow parameters, repeat cystoscopy with bladder biopsy, and MRI were employed during the 16-month follow-up period. There were no postprocedural complications. The patient reported expected symptoms of self-limited dysuria, hematuria, and flank bruising in the 2 weeks following the procedure. Urinary symptoms and flow parameters remained stable for 6 months but progressed slightly at 15 months. A repeat cystoscopy with biopsy showed stabilization of disease without concern for secondary malignancy. The patient’s 12-month postprocedural MRI and 15-month urodynamics study were unchanged from prior. This study demonstrates the safety and feasibility of adipose harvest under local anesthesia, followed by point-of-care isolation and administration of an adipose-derived cell product via cystoscopic-guided intravesical injection. This pilot is foundational for further clinical studies to elucidate effective dosing and patient selection in the treatment of bladder fibrosis with cell therapy.

Significance Statement.

This first-in-human clinical protocol examined the safety and feasibility of adipose-derived stromal vascular fraction cells for the treatment of symptomatic radiation-induced bladder fibrosis. We found that point-of-care lipoharvest, cell processing, and intravesical injection were safe and well-tolerated by the patient. While we did not observe significant clinical or histologic effects from our pilot intervention, this work serves as a foundation for further study of adipose-derived cell products in the regeneration of bladder function, establishing a reproducible clinical model in which variables of cell dosing and injection pattern can be tested in different patients.

Introduction

Pelvic radiotherapy is often administered for the treatment of colorectal, genitourinary, or gynecologic malignancies. Radiation cystitis, a clinical term describing urinary frequency, urgency, suprapubic pain, and recurrent hematuria, is a known off-target effect of both external-beam radiation therapy and brachytherapy.¹

Pathologic changes in the bladder following radiation exposure include loss of the superficial urothelial cell and glycosaminoglycan layers, persistent tissue edema, and stromal cell atypia.² Consistent with extraurinary radiation-induced tissue damage, irradiated bladders experience persistent activation of pro-inflammatory cytokines such as COX-2, leading to increased NFkB signaling.³ Fibroblast differentiation into myofibroblasts due to increased TGF-β1 activation leads to collagen accumulation and loss of smooth muscle, resulting in a loss of bladder compliance and volume.⁴ Disruption of the microvasculature also leads to abnormal proliferation of periluminal blood vessels.⁵

Treatment of radiation cystitis is primarily centered around symptom control rather than reversal of adverse pathologic changes. If behavioral modifications fail to achieve symptomatic benefit, medications such as antimuscarinics or β3 agonists are utilized to decrease overactivity of the bladder, although efficacy is often limited.^6,7 In patients with refractory hematuria, hyperbaric oxygen can promote angiogenesis and may reduce fibrosis.⁸

Adipose-derived mesenchymal stem cells (AD-MSCs) and related cellular therapies have been investigated as a disease-modifying treatment option for radiation cystitis. AD-MSCs are potent immunomodulators that reduce the secretion of proinflammatory cytokines and inhibit the differentiation of fibroblasts into myofibroblasts.⁹ The utility of MSC therapy in radiation-induced bladder injury has been demonstrated in multiple preclinical animal studies.^10-12 In humans, AD-MSC treatment has demonstrated therapeutic efficacy in multiple organ systems without adverse effects or increased rates of malignancy.^13,14 Further, injectable adipose therapies are being used as a reconstructive method for radiated soft tissues and skin across the body.¹⁵

A multitude of MSC-containing adipose cellular and tissue preparations have been previously studied. Stromal vascular fraction (SVF) cells isolated from adipose tissue containing AD-MSCs, perivascular cells, and immune cells represent a minimally modified autologous cell product that harbors significant regenerative potential.^16,17 Given the simple lipoharvest and preparation protocol, SVF cells represent a promising regenerative option amenable for point-of-care harvest and delivery.^18,19 Here, we establish the feasibility and safety of intravesical SVC injection for the treatment of radiation cystitis using a first-in-human treatment protocol to provide a foundation for future study of clinical efficacy.

Materials and methods

Preprocedural workup

An 82-year-old male with a history of prostate cancer treated with low-dose-rate brachytherapy 12 years prior presented with urinary frequency, urgency, and mixed incontinence refractory to medical therapy. A standard workup, including urodynamics, flexible cystoscopy with biopsy, and pelvic MRI with and without intravenous gadopiclenol contrast, was conducted prior to the experimental procedure. FDA approval of an extension of IDE 17634 (GID Bio) with IRB concurrence was obtained, and principles consistent with the Declaration of Helsinki were observed throughout the study.

Adipose tissue lipoaspiration

Informed consent was obtained. Saline solution with local anesthetic was instilled into the subcutaneous tissues of each flank through 1 cm incisions, and lipo-aspiration cannulas measuring 4 and 5 mm were used to collect 80 cc of lipoaspirate into a sterile GID Bio collection device. This collection volume was selected based on previously described and well-tolerated lipoharvest techniques employed for use with the processing system utilized herein.

Stromal vascular fraction (SVF) preparation

Point-of-care processing was performed using the GID Bio SVF-2 device. The lipoaspirate was triple-washed with sterile lactated Ringer’s solution within the closed system, and debris was suctioned with each wash. 5 mL of resuspended GIDzyme was mixed with the preparation at 41°C for 50 minutes. The enzymatic digestion product was centrifuged at 600 g for 6 minutes, mixed for 15 seconds, and again centrifuged at 600 g for 4 minutes. Aliquots were dispensed to evaluate viability, cell count, microscopic photography, and the presence of endotoxin.

Cystoscopy and AD-MSC injection

A flexible cystoscope was introduced into the patient’s bladder following instillation of 10 cc of lidocaine-infused lubricant. Twenty 1-cc aliquots of the resuspended SVF cells were injected into the detrusor muscle at a depth of 3 mm using the 23-ga InjecTak^® cystoscopic injection needle (Laborie). A pattern focusing on the bladder base and proximal lateral walls was used. The injection volume and quantity were selected to mimic that commonly used for chemodenervation injections to the bladder, which are routinely performed in awake patients through a flexible cystoscope. These parameters maximized the cell count we were able to safely deliver through the injection needle when accounting for total yield from the selected harvest volume. No coagulation was necessary for hemostasis. The patient was monitored for one hour after the procedure before discharge.

Clinical follow-up and monitoring

The patient was seen for an assessment of symptoms, urinary function, and physical exam seven days following the injection procedure. Uroflowmetry was performed 3, 6, 12, and 15 months after the procedure. The International Prostate Symptom Score (IPSS) questionnaire was used to assess urinary symptoms at these visits. Repeat MR with and without contrast, cystoscopy, and bladder biopsy were repeated at 12 months, and urodynamic studies were repeated 15 months after the procedure.

Results

Preoperative workup

The patient’s preprocedural uroflowmetry revealed a maximum flow rate of 11 cc/s, an average flow rate of 7 cc/s, and a voided volume of 137 cc with a post-void residual volume of 150 cc/s (Table 1). Urodynamics testing revealed detrusor overactivity with preserved bladder contractility and compliance. Preprocedural cystoscopy revealed a trabeculated bladder, pale mucosa, and mildly prominent superficial venous hypervascularity consistent with radiation cystitis. Biopsies from the bladder neck and base revealed mild lamina propria fibrosis, congestion, and chronic inflammation. A preoperative pelvic MRI revealed trabeculation of the urinary bladder, small diverticula, and no concerning perivesical soft tissue lesions.

Table 1.

Uroflowmetry parameters before and after SVF injection.

	Baseline	1 week	3 months	6 months	12 months	15 months
Max flow (cc/s)	11	10	31.2	6.5	10.2	11.2
Mean flow (cc/s)	7	7	11	3	7	8.1
Void volume (cc)	137	150	230	81	313	210
Post-void residual (cc)	167	0	100	70	110	218
Bladder volume (cc)	304	150	330	151	423	428

Cell harvest, processing, and injection

Lipoaspiration, processing, and injection were performed without complication. The total extracted cell count was 70.4 million with 73.4% viability. The samples cleared safety assays and met release criteria prior to resuspension, with a final cell count of 45.8 million. The time from initiation of lipoharvest to conclusion of injection was approximately 150 minutes. The patient’s level of discomfort remained tolerable and consistent with established office-based bladder injection procedures. The patient’s vitals remained stable for one hour, and he was discharged home in good condition. Following the procedure, the patient experienced mild hematuria that resolved spontaneously. Flank ecchymosis was noted bilaterally at the harvest sites without associated laboratory abnormalities or significant hematoma. Uroflowmetry on postprocedural day seven revealed preserved bladder function.

Urinary function outcomes

The patient’s uroflow parameters were measured at 3, 6, 12, and 15 months (Table 1). Significant fluctuation was seen between visits, with a slight trend towards larger total bladder volumes. The patient’s repeat urodynamics study revealed persistent detrusor overactivity and a stable residual volume. The patient’s responses to the IPSS questionnaire were stable over the first 12 postprocedural months with gradual worsening thereafter (Table 2). The patient had no urinary tract infections or episodes of acute urinary retention during the follow-up period.

Table 2.

Patient-reported urinary symptoms, as measured by the International Prostate Symptoms Score (IPSS) questionnaire.

	Baseline	3 months	6 months	12 months	15 months
Incomplete emptying	1	2	3	3	0
Frequency	3	3	3	2	3
Intermittency	2	2	2	3	3
Urgency	5	5	4	3	5
Weak stream	2	2	2	2	4
Straining	1	1	1	1	3
Nocturia	5	5	5	5	5
Total	19	20	20	19	23
Quality	4	4	4	4	5

Histologic and imaging outcomes

An MRI obtained 12 months after the index procedure revealed stability from the preprocedural imaging. Repeat cystoscopy with bladder biopsy revealed stability of the bladder appearance without luminal masses. Histologic examination of the biopsy specimens revealed stable chronic inflammation without evidence of dysplasia.

Discussion

Radiation cystitis is a common and often debilitating condition experienced by many survivors of pelvic malignancy that often fails to respond to existing overactive bladder therapy. The multidisciplinary protocol described herein allowed for point-of-care tissue harvest, preparation, and cystoscopic grafting under local anesthesia in a single outpatient office visit. No unanticipated postprocedural complications related to the procedure were seen. The safety of reliability of the device used and the resultant cell product have been demonstrated in a phase IIb randomized controlled trial for knee osteoarthritis with reliable rates of release criteria clearance and without significant adverse events; a phase III trial for this indication is ongoing.²⁰

In the 16 months following the procedure, the patient did not experience urinary tract infections, episodes of acute urinary retention, or gross hematuria. Comparison of the pre-and post-operative biopsy and radiographic findings demonstrated no evidence of malignancy at 12 months, supporting the safety of intravesical injection of autologous adipose products.

Although the patient’s self-reported urinary symptoms stabilized in the 12 months following treatment, ongoing progression of symptoms was noted at 15 months. These changes were consistent with the patient’s known progressive bladder fibrosis. His postprocedural uroflowmetry parameters displayed a significant variability between visits, which is common in patients with radiation fibrosis. The patient’s urodynamics findings were largely stable when accounting for expected variation between measurements.

There are potential reasons for the lack of observed improvement in symptoms in this pilot case. The histologic characteristics of this patient’s bladder tissue and time elapsed from radiation damage may limit the regenerative potential of cell-based therapies. Preclinical rodent trials utilized protocols in which MSCs were administered immediately following radiation exposure, allowing for the prevention (rather than reversal) of adverse pathologic changes.

Additionally, cell dosing and preparation type are important parameters in cell therapy treatments. Our dosing of 4.58 × 10⁷ SVF cells may not be sufficient for the surface area injected. Assuming MSCs comprised, at most, 40% of the cell count of the SVF isolate, this represents just under 1 million cells per injection site, consistent with counts previously trialed in preclinical dermal fibrosis models. While this number was selected with a focus on safety based on existing data, increased delivery volumes or utilization of purified AD-MSCs with or without cellular exosomes may elicit a more robust histopathologic and clinical response within established injection number and volume constraints.

Further, injection into fibrotic bladder tissue may result in suboptimal microanatomic engraftment of the regenerative cell population and/or limit diffusion of beneficial cellular exosomes. Repeat dosing may allow for progressive improvement of bladder histology and enhanced clinical efficacy. Alternatively, selective intraarterial cell product administration may lead to improved dispersion of the regenerative material and would be technically similar to the prostatic artery embolization procedure commonly performed by interventional radiologists.

Moving forward, the optimum patient population that may benefit from regenerative cell-based regenerative urologic treatment must be more clearly defined. Exploration of endoscopic or angiographic delivery protocols may also be necessary to unlock the therapeutic potential of genitourinary AD-MSC therapy.²¹

Conclusion

Regenerative therapy utilizing autologous adipose cell and tissue products is an emerging treatment modality across multiple anatomic systems. Preclinical data suggests this approach may hold promise for the treatment of urinary system fibrosis. We demonstrate that point-of-care harvest and intravesical cell product administration are feasible and safe in a human patient under local anesthesia, in a minor treatment room, and at a relatively low cost of care. This work lays the foundation for further clinical investigation of cell dosing, timing and frequency of interventions, and patient selection for the treatment of radiation-induced bladder fibrosis.

Author contributions

Roger Klein: Conception and design, collection and assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript. Nerone Douglas: Provision of study materials, collection and assembly of data, data analysis and interpretation, final approval of manuscript. Asim Ejaz: Conception and design, provision of study materials, collection and assembly of data, data analysis and interpretation, final approval of manuscript. J. Peter Rubin: Conception and design, financial support, administrative support, provision of study material, collection and assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript. Paul Rusilko: Conception and design, provision of patient, collection and assembly of data, data analysis and interpretation, final approval of manuscript.

Funding

J. Peter Rubin, MD, MBA served as a site investigator for clinical trial conducted by the GID Bio Company at the UPMC Health System as part of a sponsored research agreement to the health system. He is not a consultant to the company and has no financial interest in the company.

Conflicts of interest

None declared.