Treatment of vocal fold scarring with autologous bone marrow-derived human mesenchymal stromal cells-first phase I/II human clinical study

Abstract

Background: Vocal fold (VF) scarring, caused by surgery or inflammation, often results in severe voice problems or aphonia. Effective lasting treatment is lacking. Previous in vitro and in vivo animal studies reported positive effects on VF scar resolution with mesenchymal stromal cell (MSC) implantation. The principal aim of this study was to examine safety aspects and secondly treatment efficacy vocal fold function in patients with VF scarring and severe voice problems.

Methods: In this open-label phase I/II study, 16 patients were treated with surgical scar resection followed by injection of autologous MSCs (0.5-2 × 106 MSCs/patient). Patients were monitored 1 year for serious adverse events (SAE) or minor complications. Therapeutic efficacy on treated VFs was evaluated by measurement of VF vibrations using high-speed laryngoscopy (HSL) and phonation pressure threshold (PTP) for elasticity and VF function. Patients self-reported voice change using the Voice Handicap Index (VHI).

Results: No SAE or minor side effects were reported. Video ratings of VF vibrations and digitized analysis of HSL and PTP were significantly improved for 62-75% of the patients (depending on parameter). Two patients showed deteriorated VF vibrations, but improved PTP. VHI was significantly improved in 8 patients, with the remaining experiencing no significant change.

Conclusions: The results indicate that local injection of autologous MSC into scarred VFs with severe voice problems may offer a safe and feasible therapeutic option. VF vibration and elasticity were improved in approximately two thirds of treated patients. This clinical study is registered in clinicaltrials.gov (ID: NCT01981330). Retrospective registration of first patient (20130511). https//: register.clinicaltrials.gov/.

Keywords: Fibrosis; Hoarseness; Immunomodulation; Mesenchymal stromal cells; Scarring; Vocal fold; Wound healing.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

Surgery. Schematic drawing of operative technique. 1 Preoperative status with scar at vocal fold edge. 2 After cordotomy with microflap technique, scar resection. 3 Injection of MSCs in lamina propria and superficial thyroarytenoid muscle (not shown in figure). 4 Directly after surgery

Fig. 2

Vocal fold analysis of high-speed laryngoscopy. a (Top) Edge tracking of glottal area during vibration. Dots at midpoint of left and right vocal folds. Arrow marks length of membranous vocal fold part use for normalization of vibration and glottal area. b (Mid) Preoperative recording. c (Bottom) Postoperative recording for patient 12 with corresponding kymograms from the horizontal yellow line plane (left). Right vocal fold vibrations are shown above and left local fold below. Red vertical line at preoperative kymogram corresponds to maximum glottal closure (left image). High-Speed Studio software automatically sets glottal midline (red horizontal) and analyzes the brighter pixels at the most closed phase during each vibratory cycle in relation to the darker pixels during the open phase. Open/closed coefficient is calculated from this relation. Preoperative O/C coefficients in the figure are 73% preoperatively and 52% postoperatively indicating improved glottal closure. Preoperative, there is a time phase delay of maximum closure for the right vocal fold in comparison with the left vocal fold which is normalized after treatment of the right vocal fold

Fig. 3

Vocal fold recordings and analysis. (3a, 3b) Patient 1 with small scar defect on left vocal fold preoperative and 1 year postoperative (marked with asterisk). (3c) Patient 12, preoperative a small scar defect on right vocal fold (marked with two asterisks). (3d) Eight months after MSC treatment with restored vocal fold edge. (3e, 3f) Kymograms for patient 1 of vocal fold vibrations preoperative (top) and 12 months postoperative (bottom). Time scale to right. (3e) Incomplete glottal closure and reduced vibrations of the left vocal fold (lower) than right vocal fold (top). (3f) Complete glottal closure at 12 month follow-up. (3g, 3h) Kymograms for patient 12 preoperative with incomplete glottal closure and reduced vibrations (top) and 8 months postoperative with complete closure and more symmetric vocal fold vibrations (bottom). (3i) Patient 6 who has a larger defect anterior on the right vocal fold (arrow)

Fig. 4

Computerized analysis of vocal fold vibrations and phonation threshold pressure (PTP) results. Results presented as univariate plots pairwise before and 1 year after MSC treatment for each patient (maximum 16 observations before and after treatment). PTP results (d) for all 16 patients. Results for glottal minimum area (b) for 14 patients (2 patients were only examined with videostroboscopy and not with high-speed camera), open/closed coefficient (c) for 12 patients (2 were not examined with high-speed camera and for 2 patients the automatic analysis failed), glottal area variations (a) for 13 patients (2 were not examined with high-speed camera and for 1 patient the automatic analysis failed)