Inflammatory myofibroblastic tumor (IMT) is a rare mesenchymal neoplasm of intermediate malignant potential, predominantly affecting children and adolescents (1). Histologically, it is defined by a proliferation of myofibroblastic spindle cells admixed with a variable inflammatory infiltrate. The 2020 WHO classification of soft-tissue tumors recognizes IMT as an intermediate-grade lesion with a propensity for local recurrence and, rarely, distant metastasis (2). IMTs can affect multiple sites throughout the body, with pulmonary involvement being more common in adults, while extra-pulmonary IMT is more frequently seen in younger patients (3). In children, IMT most frequently involves the abdomen and pelvis, followed by the head and neck, and thoracic regions. Additionally, there have been reported cases involving the genitourinary tract, peripheral soft tissues, bones, and the central nervous system (4–6). The clinical manifestations of pediatric IMT are highly variable, primarily depending on the size and location of the tumor. The absence of specific symptoms, signs, and typical systemic or imaging features makes it challenging to distinguish IMT from other tumors clinically, leading to frequent misdiagnoses prior to surgery (7, 8).

Owing to the low incidence of pediatric IMT, the existing body of research predominantly comprises case reports or small-sample studies, with comprehensive investigations into its clinicopathologic characteristics, treatment strategies, and prognosis remaining relatively sparse. This study retrospectively analyzed the clinical data of 16 pediatric IMT patients at our hospital, aiming to summarize the clinical and pathological features, treatment methods, and prognosis, and to provide a deeper understanding of this rare disease in children, helping guide its standardized diagnosis and treatment in clinical practice.

IMT is a rare mesenchymal neoplasm of intermediate malignant potential, histologically comprising bland spindle-shaped myofibroblasts within a variable chronic inflammatory milieu (1, 9). Before being formally designated as IMT, it was referred to by various terms, including inflammatory pseudotumor, plasma cell granuloma, fibrous xanthoma, myxoid hamartoma, and benign myofibroblastic tumor, among others (10). Despite shared histology, IMT exhibits unpredictable biology: most pursue an indolent course, yet intra-abdominal and pediatric lesions display higher rates of local recurrence and, rarely, metastasis (3, 11). Precise etiology remains elusive; proposed triggers include antecedent trauma, infection, prior therapy, or germline predisposition, all potentially inciting uncontrolled myofibroblast proliferation (12–14). The present study seeks to delineate clinicopathologic correlates and outcome determinants in pediatric IMT, where such nuances directly inform risk-adapted management.

Pediatric IMT is typically diagnosed within the first two decades of life (15). In our cohort, ages ranged from 4 months to 15 years (median 5.0 years), with a modest male preponderance (M:F = 1.7:1) that most likely reflects limited sample size rather than a true sex bias. The disease can occur in various anatomical locations, but the abdominal and pelvic organs are the most common sites, followed by the head and neck, lungs, trunk, and limbs (3, 6, 16). Consistent with prior reports, abdominal and pelvic locations were most frequent in our cohort, often presenting with nonspecific gastrointestinal symptoms, including abdominal distension, pain, vomiting, fever, and palpable abdominal masses (17, 18). Head and neck lesions typically manifested as slow-growing masses (15, 19), whereas pulmonary involvement mimicked chronic infection (8, 20). Additionally, systemic inflammatory manifestations—fever, weight loss and laboratory abnormalities—are reported in 15%–30% of children with IMT (21, 22). Our findings corroborate this pattern: on admission, 43.8% of patients had leukocytosis, 56.3% had anaemia, 41.7% showed elevated CRP and 60.0% had raised ESR, whereas conventional tumour markers remained within normal limits. Importantly, these inflammatory indices-and their associated constitutional symptoms—reverted to normal within three months of complete tumour resection. This rapid postoperative normalization supports their use as simple, cost-effective surrogate markers for early detection of recurrence in settings where molecular surveillance is unavailable.

Ultrasound remains the first-line modality in children because it is rapid, non-invasive and does not require sedation (23). In the present cohort it consistently demonstrated hypoechoic solid masses with either circumscribed or infiltrative margins and scant internal vascularity, findings that were concordant with the mesenteric IMT series described by Qian et al. (23). Nevertheless, these features overlap with those of other pediatric soft-tissue neoplasms, and none of our lesions could be confidently classified pre-operatively.

In the imaging evaluation of pediatric IMT, both CT and MRI offer distinct advantages, particularly MRI, which provides superior clarity in demonstrating the relationship between the tumor and surrounding soft tissues, thereby offering valuable information for treatment planning and prognostic assessment (24). On contrast-enhanced CT the tumours were heterogeneously attenuated, with variable patterns of enhancement (peripheral, progressive, or delayed) and occasional central necrosis or punctate calcification, as previously reported (24). On MRI scans, most lesions appear as soft tissue masses, with T1-weighted imaging (T1WI) showing isointense or hypointense signals and T2-weighted imaging (T2WI) showing isointense or hyperintense signals. The signal characteristics are closely related to the degree of fibrosis and inflammatory cell infiltration within the lesion. Contrast-enhanced MRI demonstrates a variety of enhancement patterns (24, 25). However, neither modality permitted reliable differentiation from malignant sarcoma or lymphoma.

Endoscopic evaluation (laryngoscopy, bronchoscopy or gastrointestinal endoscopy) was invaluable for submucosal lesions; in two cases it revealed smooth polypoid masses that were subsequently excised endoscopically (26). 18F-FDG PET-CT was not routinely employed, reflecting its recognized limitation in IMT: variable tracer uptake related to cellularity and inflammation frequently results in false-positive results (27). Previous studies have reported cases where IMT was misdiagnosed as lymphoma on PET-CT (28). While PET-CT cannot reliably diagnose IMT due to its variable FDG uptake, it remains a valuable tool for detecting primary tumors, assessing local recurrence and distant metastasis, as well as evaluating treatment response (29).

In our study, ultrasound was performed in 13 children, revealing hypoechoic solid masses with well-defined margins. Particularly in cases of abdominal IMT, ultrasound effectively diagnosed complications such as intestinal intussusception and bowel obstruction. However, it remained challenging to definitively characterize the nature of the lesions preoperatively. Nine patients underwent CT scans, and two patients underwent MRI scans, but the imaging findings lacked specificity and failed to provide a correct preoperative diagnosis. This highlights that pediatric IMTs exhibit diverse imaging features, some of which may mimic malignancy. However, relying solely on preoperative imaging for a definitive diagnosis remains difficult.

Currently, pathological examination and IHC remain the “gold standard” for diagnosing IMT, with ALK gene rearrangement and ALK protein expression serving as crucial diagnostic criteria (9, 30). Microscopically, the tumour is composed of cytologically bland, spindle-shaped myofibroblasts arranged in fascicles, accompanied by a dense mixed inflammatory infiltrate rich in lymphocytes, plasma cells and eosinophils (31). IHC plays a critical role in the diagnosis of IMT by identifying the immunophenotype of myofibroblasts and excluding other similar diseases (30). In IMT, SMA and vimentin are typically positiven (32), desmin may be focally expressed, whereas S-100, CD34 and CD117 are consistently negative (31, 33). ALK is particularly significant for the diagnosis of IMT due to its relatively high sensitivity and specificity. It is also recognized as one of the driver genes of IMT (34). ALK immunoreactivity was detected in 62.5% (10/16) of our cases; SMA, vimentin and desmin positivity were observed in 80.0% (12/15), 78.6% (11/14) and 63.7% (7/11), respectively.

Owing to insurance constraints, FISH for ALK rearrangement was performed in only three patients, yielding two positive and one negative result, consistent with prior series reporting ALK rearrangements in 50%–70% of cases (33). The ALK gene may fuse with multiple partner genes, such as TPM3/4, GCC2, TRAF3, EML4, and THBS1, which could be one of the key mechanisms underlying the development of IMT (35–37). In ALK-negative tumours, alternative drivers have been identified, including ROS1, RET, NTRK3 and PDGFRB rearrangements or mutations (32, 38, 39). These alterations define a genomically distinct subset that may respond to corresponding TKIs (e.g., entrectinib for NTRK3, selpercatinib for RET, avapritinib for PDGFRB) (39, 40). Although our cohort was not systematically profiled for these drivers, we acknowledge that prospective protocols should incorporate comprehensive molecular panels to capture actionable fusions and to refine risk-adapted therapy in ALK-negative cases.

Although malignant transformation is reported in 8%–18% of IMTs, characterised by loss of spindle morphology, increased mitoses and necrosis (41), none of our three recurrent specimens displayed evidence of high-grade progression. This may reflect the indolent biology of pediatric lesions, early detection of relapse, or the small number of recurrences studied. Serial sampling of recurrent or metastatic foci is therefore warranted to capture any evolution toward aggressive disease (42, 43).

Pre-operative diagnosis can be challenging. Image-guided core-needle or endoscopic biopsy is often attempted for deep-seated lesions, but limited material, crush artefacts and dense inflammation frequently yield inconclusive results (44). In a recent pediatric series, initial biopsies failed to establish the diagnosis in 3 of 4 IMT cases, mandating repeat sampling or upfront resection (45). ALK-negative or IgG4-mimicking tumours are particularly prone to misclassification. Thus, when clinical-radiological suspicion remains high despite a non-diagnostic biopsy, clinicians should consider i. larger-gauge or incisional re-biopsy, ii. frozen-section-guided early excision, or iii. direct definitive surgery if morbidity is acceptable.

Contemporary guidelines establish complete surgical excision with histologically negative margins as the cornerstone of therapy for pediatric IMT (22, 46). The operative strategy is dictated by tumour site, size and local infiltration, with the primary objective of achieving R0 resection whenever anatomically feasible (9). Published series indicate that children undergoing R0 resection experience excellent long-term outcomes, with 5-year survival consistently exceeding 90% (47). Adjuvant treatment after R0 resection remains contentious. Current evidence does not support the routine addition of radiotherapy or chemotherapy in this setting, especially in view of the potential long-term sequelae of irradiation—namely growth retardation and secondary malignancy—which have led to recommendations against radiotherapy in children under three years (48). For patients with unresectable disease, incomplete excision (R1/R2) or documented recurrence, systemic options are considered (15). Corticosteroids, non-steroidal anti-inflammatory drugs and cytotoxic regimens containing anthracyclines with vincristine or vinblastine have produced objective response rates of 48%–64%, with durable disease control reported in selected cases (48, 49).

Beyond histopathology, molecular profiling now guides therapeutic decisions. ALK rearrangements are detected in more than half of pediatric IMTs, providing a rational target for precision therapy (40, 50). In advanced, unresectable or relapsed ALK-positive disease, ALK inhibition has moved from salvage to adjuvant strategy.Ceritinib achieved an objective response rate of 70% among ten heavily pre-treated children (51), while crizotinib produced complete remission in seven of fourteen pediatric patients and partial remission in five, for an overall response rate of 86% (52). These data establish ALK-directed therapy as an effective first-line option in genomically defined subsets. ALK-negative tumours lack this target and may pursue a more aggressive course, with higher metastatic risk. Empirical approaches—including NSAIDs, rituximab and alternative tyrosine-kinase inhibitors—have shown anecdotal benefit in this subgroup (53, 54), but prospective validation is required to define optimal management.

With increasing understanding of pediatric IMT, neoadjuvant therapy has gradually gained attention for its application in pediatric IMT and has demonstrated significant clinical benefits in some patients (55). Pre-operative ALK inhibition or chemotherapy can down-stage initially unresectable tumours, enabling R0 resection while maximally preserving adjacent organs (50). In a minority of cases, this approach has induced complete remission, thereby obviating the need for extensive surgery (56). Consequently, neoadjuvant therapy should be considered for lesions in anatomically challenging sites or for patients with early recurrence, although prospective protocols are required to define optimal timing, duration and patient selection.

Pulmonary IMT deserves particular mention because it represents the most frequently reported visceral site in both children and adults. According to two recently published studies, lung IMTs account for the majority of thoracic cases and are characterized by a distinctly indolent biology: complete surgical resection is associated with a recurrence rate as low as 2%, whereas incomplete resection carries up to a 60% recurrence risk (57, 58). The rarity of metastatic disease (<5%) and the excellent long-term survival after R0 resection support the current recommendation that surgery remains the cornerstone of treatment, with systemic therapy reserved for unresectable or relapsed disease (58). These data corroborate our own observation that the single pulmonary IMT in our cohort (Table 1, Case 8) has remained disease-free for 46 months after complete resection and reinforce the importance of achieving negative margins irrespective of anatomic site.

Pediatric IMT generally has a favorable prognosis, yet the recurrence rate varies with the anatomical location of the tumor. It has been reported that the recurrence rate for pulmonary IMT is approximately 2%–5%, while for extra-pulmonary IMT, the recurrence rate is significantly higher, around 25% (15, 59). Distant metastasis is relatively rare in pediatric IMT. Multivariate analyses have identified tumour location, local invasiveness, multi-focal growth, positive surgical margins, ALK expression status and specific gene fusions as independent risk factors (9, 31, 60). In particular, mesenteric or peritoneal disease with a multinodular pattern, as well as tumors situated in the pharynx, larynx or sinonasal tract, recur more frequently, presumably because complete surgical extirpation is technically challenging (61, 62). Across studies, margin status emerges as the single most robust predictor of local relapse and ultimately influences both survival and subsequent recurrence risk (48, 59, 63).

Optimal therapy for recurrent pediatric IMT has not been prospectively defined. In fit children with limited local relapse, repeat resection remains the treatment of choice, provided complete excision is achievable (49). In our cohort, all 16 patients underwent primary surgery (1 endoscopic, 15 open); 15 obtained R0 margins and none received adjuvant therapy. During a median follow-up of 41 months, three patients (18.8%) developed local recurrence and underwent second operations. Two underwent repeat R0 resections without further therapy. The third, who presented with multi-focal peritoneal disease, achieved disease stabilisation on crizotinib after incomplete debulking. Notably, the index resections of all three recurrences shared several features: one originated from an R1 endoscopic excision, and the two mesenteric primaries measured 8 cm and 10 cm, respectively. Additionally, the Ki-67 labeling index was markedly higher in recurrent lesions (mean 18.3%) than in non-recurrent tumors (8.5%), although this observation is limited by the small sample size.These data align with recent adult and pediatric series suggesting that Ki-67 > 10% may identify a subset with increased proliferative activity and a higher likelihood of local relapse (59, 64). While Ki-67 is not yet a validated prognostic biomarker in IMT, it could complement tumour site, margin status and molecular subtype in future multi-parametric risk-stratification models.

Pediatric IMT is a rare, low-grade neoplasm with an excellent prognosis after complete surgical excision. Pre-operative diagnosis remains challenging because of non-specific clinical and radiological features; histopathology and a targeted IHC panel remain indispensable. R0 resection is the cornerstone of therapy; adjuvant treatment, if required, should be individualised according to tumour site, margin status, Ki-67 index and other clinicopathological risk factors. Given the potential for recurrence and metastasis in pediatric IMT, long-term monitoring and regular follow-up are crucial for early detection of recurrence or metastatic lesions.