A Pilot Study of Intralesional Injection of Triamcinolone Acetonide for Desmoid Tumors: Two-Year Outcomes

Although benign, desmoid tumor is challenging to clinically manage due to its locally aggressive behavior and propensity for recurrence. There are several treatment options available, but each are associated with drawbacks. Surgical resection has historically been the treatment of choice but can lead to disfigurement and is associated with a 30–40% rate of local recurrence (1). More recently, nonsurgical treatments including radiation, systemic therapies, and locally ablative options have demonstrated improved response rates (2, 3). However, these therapies are also associated with undesirable side effects (4–6). Sorafenib and nirogacestat might affect fertility. Locally ablative options, such as cryotherapy, may not be suitable in certain locations such as when the tumor is located close to the skin or adjacent to critical structures such as major nerves or vessels (7–10).

Observation alone is a reasonable initial approach because studies have shown that up to 60% of desmoid tumors stabilize while up to 25% spontaneously regress over a 5-year period (11–15). Additional larger longitudinal prospective studies are needed to determine the natural history of desmoid tumors and to identify potential clinical and imaging features that could help predict prognosis. For tumors that fail observation, effective and well-tolerated treatment strategies are needed.

Desmoid tumors are histologically similar to keloids and superficial fibromatoses and are composed of proliferations of fibroblasts and myofibroblasts intermixed with abundant dense collagen (6). In keloids, the corticosteroid triamcinolone acetonide has long been effectively used to reduce lesion size (16). More recently, triamcinolone acetonide has also shown promise in the treatment of superficial fibromatoses, specifically Dupuytren's nodules (17–19). Despite the success of triamcinolone in keloids and superficial fibromatoses, there are only two reports of intralesional steroid use for treatment of desmoid tumors (20, 21).

The goal of this pilot study was to investigate serial percutaneous intralesional injections of triamcinolone acetonide for treatment of histologically confirmed extra-abdominal desmoid tumors. The primary objective was to determine the response rate based on the following: (i) World Health Organization (WHO) response criteria, (ii) RECIST, (iii) change in tumor volume on serial MRI, and (iv) change in T2 and T1 post-contrast signal intensity on serial MRI (21–23). The secondary objectives included assessment of safety and tolerability of the injections. We hypothesized that triamcinolone acetonide injections would result in decreased tumor size and that they would be safe to perform and tolerated by patients.

Following Institutional Review Board (IRB) approval and registration on clinicaltrials.gov (NCT03627741), subjects were recruited through the clinical practice of authors S. Attia and B.K. Wilke. Patients were screened for eligibility during their clinic visit, and if eligible, offered enrollment in the trial. Written informed consent was obtained and the study was conducted in accordance with the Declaration of Helsinki guidelines. For inclusion in the study, patients were required to have a histologically confirmed diagnosis of extra-abdominal aggressive fibromatosis (desmoid tumor). Patients must also have had either (i) a desmoid tumor that had shown stability in size over consecutive axial imaging at least 3 months apart and presence of tumor-related symptoms or (ii) an increase in size based on consecutive axial imaging at least 3 months apart. Patients who had received systemic therapy must have been off treatment for at least five drug half-lives prior to enrolling in the trial. For patients with a desmoid tumor that had been irradiated, at least a 10% increase in size by volume since receiving radiotherapy was required.

Patients were ineligible for the trial if they were actively undergoing concomitant treatment. Patients who were pregnant or attempting to become pregnant, and those who had an active infection or uncontrolled illness that would limit compliance with study requirements were also excluded.

The percutaneous intralesional injections were performed under ultrasound guidance by two fellowship-trained musculoskeletal radiologists (H.W. Garner and J.M. Bestic). Informed consent for the procedure was obtained by the radiologist prior to each injection. Three milliliters containing a concentration of 40 mg/mL of triamcinolone acetonide was used, for a total of 120 mg per injection. The injection was administered into the center of the tumor and three total injections were performed at 6-week intervals (0, 6, and 12 weeks). The treatment dosing and schedule were based on previous studies of triamcinolone use in keloids (7) and superficial fibromatoses (16, 18, 19).

The therapeutic response was analyzed on MRI based on the following: (i) WHO response criteria, (ii) RECIST, (iii) change in tumor volume on serial MRI, and (iv) change in T2 and T1 post-contrast signal intensity ratio on serial MRI. MRI of the tumor was obtained prior to the first injection (baseline) and compared with MRIs obtained at 6-month intervals after enrollment in the trial, out to 24 months. The tumor volume was measured in the same manner as described by Sheth and colleagues using the equation, |$Volume\ = \ \frac{1}{2} \times Diameter1 \times Diameter2 \times Diameter3$| (14). The T2 and T1 post-contrast signal intensity ratio was calculated using the method also described by Sheth and colleagues in which a region of interest (ROI) was drawn as large as possible within the tumor on both a T2 and a T1 post-contrast image where the tumor was at its largest diameter. The mean signal intensity on T2 and T1 post-contrast images was normalized to the muscle within the same field-of-view providing a tumor:muscle ratio on both T2 and T1 post-contrast images (Fig. 1). The MRI examinations were interpreted by a single musculoskeletal fellowship-trained radiologist (H.W. Garner) for consistency.

Safety and tolerability were assessed by clinical evaluation prior to proceeding with each additional injection. Adverse events and patient-reported outcomes (PRO) scores were recorded at each visit, including the 12-item short form survey (SF-12), visual analog scale (VAS), and desmoid PRO tool (24).

Continuous variables were summarized with the sample median and range. Categorical variables were summarized with number and percentage of patients. Comparisons of outcomes between the baseline, 6-month, 12-month, 18-month, and 24-month follow-ups were made using a paired Wilcoxon signed rank test (continuous and ordinal outcomes) or a paired McNemar test (binary categorical outcomes). For missing data, the last observation carried forward method was applied. P values < 0.05 were considered statistically significant and all statistical tests were two-sided. Statistical analysis was performed using R Statistical Software (version 4.1.2; R Foundation for Statistical Computing, Vienna, Austria).

Deidentified study data may be made available upon written request to the authors.

Ten patients consented to participate in the study over a 16-month period (December 2018 to April 2020). Written informed consent was obtained and the study conducted in accordance with our IRB, under the ethical guidelines proposed in the Declaration of Helsinki. One patient withdrew prior to undergoing the injections due to difficulties with transportation for follow-up appointments. Nine patients were included in the final analysis. No patient received additional desmoid-specific treatment other than the steroids during the study period.

Patient demographic information was collected at baseline (Table 1), and was representative of the broader desmoid community (Supplementary Table S1). Seven of 9 patients (77.8%) had received prior treatment for their desmoid tumor. This treatment included surgery, radiation, and systemic therapy. The last treatment received by the patients, and the time from cessation of that treatment to enrollment in this study is listed in Table 2. Four patients (44.4%) had tumors located in the lower extremity, 3 (33.3%) in the chest wall, and 3 (22.2%) in the upper extremity.

Comparisons of MRI parameters between the baseline and follow-up imaging are provided in Table 3 and Fig. 2. Of the 9 study patients, 8 (88.9%) demonstrated a reduction in the volume of their tumor while 1 (11.1%) enlarged. The median change in tumor volume between the baseline and 6-month MRI was −16.3% (−37.2% to 63.6%; P = 0.13). The median change in tumor volume between the baseline and 24-month MRI was −26.9% (−81.1% to 34.6%; P = 0.055). On the basis of WHO criteria, all tumors remained stable (100%) at the 24-month follow-up, whereas two tumors (22.2%) demonstrated partial response based on RECIST (Fig. 3).

There was a significant decrease in the mean tumor:muscle post-contrast signal intensity ratio from baseline to 6 months (P = 0.008) and from baseline to 24 months (P = 0.004). There was a similar observed decrease in the mean tumor:muscle T2 signal intensity at 24 months when compared with the baseline imaging (P = 0.02). The percentage change in the 6-month signal intensity ratio of the post-contrast imaging was negatively correlated with tumor volume changes at 24 months (Spearman's r, −0.80; P = 0.014; Supplementary Figs. S1 and S2).

Comparisons of SF-12 and VAS information between the baseline and both 6 months and 24 months were compared (Supplementary Table S2). No differences were observed between time points (all P > 0.05). Supplementary Table S3 displays comparisons of the desmoid tumor PRO. There were statistically significant improvements from baseline to both 6 months and 24 months related to difficulties with activity, fear, and anxiety (all P > 0.05).

Complications following the injections included swelling at the injection site (n = 1), bruising (n = 1), hypopigmentation (n = 1), and pain at the injection site (n = 1). One patient who was diagnosed with “adrenal fatigue” a year prior to study inclusion, was diagnosed with adrenal insufficiency approximately 1 month following the final injection and treated with a 6-week course of oral steroids, consisting of 25 mg of oral hydrocortisone, by the treating endocrinologist. This patient's tumor response at 24 months was −70%. The patient had not previously received steroids for the adrenal fatigue diagnosis.

Triamcinolone acetonide is a safe and effective treatment for keloids and hypertrophic scars. Proposed mechanisms of action include a decrease in production of collagen, dissolution of insoluble collagen (collagenolysis), a decrease in local inflammation, and an increased rate of apoptosis of fibroblasts and inflammatory cells (25, 26, 27–29). In the treatment of keloids, triamcinolone has a reported response rate of 50% to 100% (30). More recently, triamcinolone has been evaluated in the treatment of superficial palmar fibromatoses (Dupuytren's nodules) of the hand. Three reports have demonstrated regression of the nodules and increased range of motion following serial injections (17–19).

Although histologically similar to keloids and superficial fibromatoses, there are only a few reports regarding corticosteroid treatment of extra-abdominal desmoid tumors and retroperitoneal fibrosis (11, 12, 31–33). Rhee and colleagues reported a case of a chest wall desmoid tumor that recurred after two surgical resections and postoperative radiation therapy. The authors treated the lesion with weekly intralesional injections of 120 mg of triamcinolone acetonide for 4 weeks. At 6 months, they noted a reduction in the size of the tumor (20). Similarly, Umemoto and colleagues reported a case of a 37-year-old man with familial polyposis coli and intra-abdominal desmoid tumors. He was treated with oral prednisolone therapy with gradual regression of the lesions (34). More recently, Pruksakorn and colleagues reported on serial corticosteroid injections into eight desmoid tumors. They reported that the serial injections appeared to be a safe and potentially effective treatment, noting stabilization of the tumors in 83% of cases (21).

While response rates for desmoid tumors have historically been reported using WHO response criteria or RECIST, the use of these tools has been questioned due to the slow nature of response with a tumor that is composed primarily of fibroblastic tissue (11, 31, 35, 36). Investigators have instead suggested using change in tumor volume and signal intensity ratio on T2 and T1 post-contrast MRI as a surrogate of treatment efficacy (23, 37). The rationale is that highly cellular areas will demonstrate high T2 signal intensity and T1 post contrast enhancement, while quiescent areas will have lower T2 signal intensity and less T1 post contrast enhancement. It is thought that treatment effect may be observed more quickly based on a decrease in signal intensity ratio as opposed to a change in tumor size (11, 35, 38, 39). Supporting this idea, we observed a larger change in tumor volume and signal intensity ratios at 24 months than when using the standard tools for evaluating tumor response. Of interest, however, our data demonstrated an inverse correlation between the early change in tumor signal intensity and the change in tumor volume at 24 months. Larger studies will be needed to confirm or refute these initial results.

When comparing PRO data, we were unable to find statistically significant differences in SF-12 and VAS scores between the baseline, 6-month, and 24-month follow-up questionnaires. However, using the desmoid PRO tool we did find improvements in a few domains, specifically activity, fear, and anxiety. This tool was designed for desmoid patients based on their perspectives on key issues related to their disease and may be a more sensitive test for patient satisfaction with treatment (24, 40).

Previous reports have documented adverse events following percutaneous injections of triamcinolone acetonide. These included hypopigmentation and subcutaneous atrophy at the injection site as well as menorrhagia and cushingoid features. These latter events were transient and reportedly resolved within three months (17, 26). In our cohort, a single patient reported hypopigmentation at the injection site, 1 patient reported temporary pain at the injection site, and an additional patient reported bruising. No patients developed cushingoid features, but 1 patient who had a previous diagnosis of “adrenal fatigue” was diagnosed with adrenal insufficiency. This patient was treated with a 6-week course of oral steroids following the injections. In the study by Pruksakorn and colleagues, hypothalamic-pituitary-adrenal (HPA) axis suppression was noted in half of their patients, correlating with the dose of steroids given (31). Monitoring of the HPA axis would be valuable in future studies to determine optimal dosing of triamcinolone.

The main limitation of this study is the small sample size, which though appropriate for a pilot study, results in a lack of power. Therefore, the possibility of a type II error (i.e., a false-negative finding) is important to consider. For comparative variables, we cannot conclude that no true difference exists simply due to the occurrence of a nonsignificant P value in our small study.

An additional limitation is the wide variability in volumes of desmoid tumors treated. While all patients received the same dosing (120 mg/mL of triamcinolone acetonide), there was a wide range in baseline tumor volume. It is possible that a “one size fits all” approach to injection scheduling and dosing may not be appropriate. Similarly, all injections were placed into the center of the tumors. This was chosen to improve injection consistency between radiologists. It is possible that multiple injection sites throughout the tumor or more strategic placement of the injection into highly cellular areas would provide greater benefit.

Finally, genetic analysis of the desmoid tumors was not performed, thereby limiting our ability to compare the eight tumors that decreased in volume to the one tumor that increased. Larger multicenter studies are needed to further evaluate the treatment effect of serial triamcinolone injections. For next steps, we propose a randomized controlled trial comparing triamcinolone injections with active surveillance, allowing crossover in the surveillance group with tumor progression.

Treatment of desmoid tumors with serial triamcinolone injections appears safe and well tolerated by patients, with a 22% partial response rate based on RECIST. Further research is needed to confirm our results and to determine factors predictive of response.

B.K. Wilke reports grants from Desmoid Tumor Research Foundation during the conduct of the study. H.W. Garner reports grants from Desmoid Tumor Research Foundation during the conduct of the study. L.A. Chase reports grants from Desmoid Tumor Research Foundation during the conduct of the study. S. Attia reports grants from Desmoid Tumor Research Foundation during the conduct of the study, as well as grants from Springworks, Bayer, Tracon, Takeda, Novartis, Ayela, Lilly, Karyopharm, Epizyme, Blueprint, CBA, Merck, Philogen, Gradilis, Deciphera, Incyte, Adaptimmune, Bavarian Nordic, BTG, PTC, GSK, Forma, Trillium, Theseus, BI, Noxopharm, Inhibrx, Salarius, C4, and Rain outside the submitted work. No disclosures were reported by the other authors.

B.K. Wilke: Conceptualization, data curation, formal analysis, funding acquisition, methodology, writing–original draft. H.W. Garner: Data curation, writing–review and editing. J.M. Bestic: Data curation, writing–review and editing. L.A. Chase: Methodology. M.G. Heckman: Formal analysis, writing–review and editing. J.J. Schoch: Conceptualization, writing–review and editing. S. Attia: Methodology, writing–review and editing.

Funding for this project was received from the Desmoid Tumor Research Foundation. Award recipients were authors B.K. Wilke and S. Attia.

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.