Selumetinib in the treatment of plexiform neurofibromas in neurofibromatosis type 1: a 6-year prospective case series with review of the literature Authors: Francesco Baldo, Luca Messana, Alberto Andresani, Andrea Magnolato, Flora Maria Murru, Anita Spezzacatene, Elisabetta Cattaruzzi, Alessandra Maestro, Gilda Paternuosto, Davide Zanon, Laura Diplotti, Paola Michieletto, Egidio Barbi, Irene Bruno Published: 2026 ABSTRACT AbstractBackgroundNeurofibromatosis type 1 (NF1) is an autosomal dominant genetic disorder characterized by pigmented lesions, cutaneous neurofibromas, and brain and peripheral nerve tumors. Plexiform neurofibromas (PNs) are congenital tumors whose growth tends to accelerate in the pediatric age. Selumetinib, a MEK1/2 inhibitor, is the first pharmacological treatment that has shown efficacy in reducing the size of PNs by inhibiting their key pathway, with few and moderate adverse events. The purpose of this study is to describe the effect of selumetinib in a large cohort of patients with NF1.MethodsThis is a prospective case series that describes the patients affected by NF1 with inoperable, symptomatic, or disfiguring PNs treated with selumetinib at the Institute for Maternal and Child Health IRCCS “Burlo Garofolo” in Trieste, from January 2017 to June 2023. To determine the patients’ response to selumetinib, we compared the MRI volume of the PN before the beginning of the treatment with the last one available.ResultsTwenty-three patients with PNs were treated with selumetinib. During the follow-up period (mean time 1085 days), a tumor reduction above 20% was observed in 13 patients (57%), while 7 subjects (30%) had a tumor stabilization and 3 (13%) had a tumor growth. The median volumetric reduction was 23%. No significant correlations were found between the patients’ response to treatment and the other variables considered. No severe adverse events were recorded.ConclusionsIn this prospective case series selumetinib appeared to be a useful drug for the treatment of PNs.Keywords: Neurofibromatosis, Selumetinib, Plexiform neurofibroma, NF1 Background Neurofibromatosis type 1 (NF1) is an autosomal dominant genetic disorder caused by mutations in the NF1 gene [1]. Individuals with NF1 typically present with pigmented lesions, that usually manifest in childhood (café au lait spots, axillary and inguinal freckling, irideal Lisch nodules), and multiple cutaneous neurofibromas, that normally appear later in adulthood. Skeletal abnormalities (including tibial dysplasia and scoliosis), vasculopathy, hypertension, and learning disabilities are also typical of this condition. Nonetheless, the most relevant feature of NF1, and therefore the core of its clinical management and follow-up, is the risk of developing brain tumors (optic nerve and central nervous system gliomas) and peripheral nerve tumors (plexiform neurofibromas and malignant peripheral nerve sheath tumors) [2]. Of the latter group, plexiform neurofibromas (PNs) are the most frequent. It is estimated that about 50% of the individuals affected by NF1 will develop one PN during their lifetime [3]. PNs are congenital, but their growth tends to accelerate in the pediatric age, reaching a maximum in the prepubertal phase and usually stopping at the end of the pubertal spurt [4, 5]. Although histologically benign, PNs can cause pain and neurological deficits, movements’ limitation, and disfigurement, and they can compress internal organs (including the upper and lower airways) and blood vessels [6]. Furthermore, especially when large and extensive, they might evolve into malignant peripheral nerve sheath tumors (MPNST). In the past, surgery has been the primary therapeutic option for PNs. However, since a complete resection is often hard to achieve and tumor remnants may result in a regrowth of the mass, this approach often offers poor results [7]. Therefore, a significant emphasis has progressively been put on identifying a pharmacological treatment that could reduce the size of the tumors, possibly avoiding surgical procedures. Selumetinib is a small molecule that acts as an ATP-independent inhibitor of mitogen-activated protein kinase (MEK or MAPK/ERK kinase) 1 and 2. MEK 1 and 2 are key mediators in the activation of the RAS/RAF/MEK/ERK pathway, which is upregulated in NF1 due to the altered function of the tumor suppressor neurofibromin that usually inhibits this pathway. This hyperactivation increases the production of transcription factors, and causes cell survival and proliferation, leading to the typical clinical features of NF1. In 2016, Dombi et al. conducted the first phase I trial with selumetinib in the pediatric population, with positive results [8]. In fact, a tumor volume regression from baseline ≥ 20% (i.e., partial response) was achieved in 17 out of 24 children (71%) with inoperable PNs, and none of the others presented disease progression. A decrease in tumor-related pain, disfigurement, and functional impairment was also observed. Most of the adverse events observed were mild, and included gastrointestinal symptoms (especially diarrhea), maculopapular or acneiform rash (in pre-pubertal and post-pubertal patients, respectively), and asymptomatic elevation of the creatine kinase. These positive data were confirmed by the following phase II study, in which 70% of patients had a durable tumor shrinkage, along with noticeable clinical benefits, including reduced tumor pain intensity, improved strength and range of motion, and increased quality of life [9]. In consideration of these positive results, since November 2017 at the Institute of Maternal and Child Health Burlo Garofolo in Trieste, Italy, we started to administer selumetinib for compassionate use to patients with NF1 affected by inoperable, symptomatic, or disfiguring PNs. Our group’s experience was the first one reported in Italy and the second in Europe and confirmed the positive effects of this drug in our pediatric cohort (clinical and radiological response in 8 out of 9 patients) [10]. In April 2020, the FDA (Food and Drug administration) approved selumetinib for pediatric patients. One year later the EMA (European Medicines Agency) did the same and eventually, in January 2024, the Italian Medicines Agency (AIFA) approved selumetinib redeemability in Italy. Since the number of patients treated with selumetinib in our center has been steadily growing, we considered necessary to reanalyze our cohort data, to increase the data available for the medical community, both in terms of efficacy and safety profile. Methods This is a prospective case series that describes the patients affected by NF1 with inoperable, symptomatic, or disfiguring PNs treated with selumetinib at the Institute for Maternal and Child Health IRCCS “Burlo Garofolo” in Trieste, from January 2017 to June 2023. Selumetinib was accessed through compassionate use. All the patients with PNs evaluated at least twice with MRI (one prior to the treatment, and one after at least six months after the beginning of the treatment) were included. For each patient we considered the following variables: age (at the beginning of the treatment and current), gender, number and localization of the PNs, presence of an optic glioma, duration of the treatment and possible dose modification and/or interruption. All the patients were treated with 25 mg/sqm of selumetinib twice day, as currently recommended. The patients have been monitored closely during the follow up period with periodic visits (every three to four months in the first year, and then every six months). Routine checks included general/pediatric evaluation made by physicians with expertise in NF1, a cardiological evaluation with EKG and echocardiography, an ophthalmological evaluation, a pneumological evaluation with spirometry, and blood exams. The latter included full blood count, AST (aspartate aminotransferase), ALT (alanine aminotransferase), CPK (creatin phosphokinase), electrolytes (sodium, potassium, chloride, calcium, phosphorus, and magnesium), albumin, protein electrophoresis, IgA, IgM, and IgG. In case of emergency, the patients’ family could directly contact one of the physicians, to communicate any adverse event and receive clinical advice for its treatment, or to ask for a prompt clinical revaluation. MRIs to determine the volume of the PNs were performed before the treatment and then every 6 months to 9 months. To determine the patients’ response to selumetinib, we compared the PN’s volume at MRI before the beginning of the treatment with the last one available. If a patient had multiple neurofibromas, we chose the most clinically relevant. To determine the volume of the PNs a 3D evaluation with the software Horos™ was performed. This choice was motivated by the fact that the neurofibromas’ shape is typically irregular, and a volume determination calculated using only the three axes would have not been accurate. The volumetric evaluation was performed on STIR or T2 weighted sequences by a group of radiologists with expertise in NF1 with the support of pediatricians. The difference between the volumes at the first and last MRI was expressed as a percentage. As previously done by Dombi et al. we divided the volumetric responses in three different groups: Regression: every reduction above 20% Regression: every reduction above 20% Progression: every increase above 20% Progression: every increase above 20% Stabilization: every size change below 20% Stabilization: every size change below 20% Informed consent was signed by parents at the first visit, in which they agreed that “clinical data may be used for clinical research purposes, epidemiology, the study of pathologies and training, to improve knowledge, care and prevention” [11]. All data were collected in an anonymous database. All statistical analyses were conducted with JMP™ (version 16.1.0, SAS Institute Inc., Cary, NC, United States). Descriptive statistics was used to describe data. Continuous variables were expressed as mean, median, minimum, and maximum. Categorical data were expressed as percentages (%). For the comparison of continuous variables between more than two groups, the Kruskal-Wallis test was used. The chi-square test was used to compare categorical variables between the groups. To compare our results with those reported in the current medical literature, we researched on Pubmed using the words “selumetinib”, “treatment”, “plexiform neurofibromas”, “children” and “pediatrics”, Results Twenty-three patients with PNs were treated with selumetinib. All of them had at least one progressive PN, whose growth was determined radiologically (previous MRIs and/or ultrasound, depending on the area involved), and clinically (physical evaluation) when assessable. None of them was medically treated before and during the treatment with selumetinib with any other drug targeting PNs growth. The symptom most frequently reported was pain (56%). Other reasons to start the treatment were the localization next to a potentially critical site (26%) and the presence of a disfiguring lesion (17%). One patient with head localization developed dysphagia, which resolved during the treatment. The summary of the patients’ demographic features is reported in Table 1. All the subjects started the treatment before reaching 18 years of age, except for three, who were respectively 18, 19 and 36 years of age at first administration. As for the follow-up period, mean time was 1085 days and median time 1007 days (minimum 318 days, maximum 2060 days). Overall, we identified 49 PNs. Their distribution was heterogenous: 16 of them were in the head and neck region, 4 in the upper limbs, 9 in the lower limbs, 11 in the thorax and back area, and 9 in the abdominal and pelvic region. The median number of PNs per patient was 2 (maximum 4, minimum 1). In 6 patients (26%) an optic pathway glioma was also present. The patient baseline evaluation (including PN volume before the treatment and at latest follow up) is reported in Table 2. In the patients who had a tumor stabilization or reduction, the best volumetric response was always observed in the last MRI evaluated. In the two patients that had a tumor progression, the PN initially shrank partially (PN stabilization) and after two years of treatment the lesion started to enlarge. The summary of the response to the treatment is reported in Table 3. We could not find any significant correlation between the patients’ response and all the other variables considered (age, PN localization, main symptom, length of treatment). Table 1Demographic features of the patients treated with selumetinibPatients23 (100%) < 18 years of age20 (87%) > 18 years of age3 (13%)Age at the start of the treatment Mean12 Median12 Minimum3 Maximum36Current age Mean15 Median15 Minimum7 Maximum37Male14 (61%)Female9 (39%) Demographic features of the patients treated with selumetinib Table 2Baseline evaluation of the patients: PN evaluated and correlated issue, interval between the pre-treatment MRI and the last one (days), length of the treatment (days), PN volume (cm3) before the treatment and at latest MRI evaluation, and volume change expressed as percentage; in the latter, (+) indicates a volumetric increasePatientSexAge PN localizationMain issueMRI to treatment (weeks)Treatment (days)PN before PN afterMRI interval (days)Volume change (%)1M13ParavertebralLocalization*113581611101069322M12FootPain120604893091855373M13LegPain131675746701149664F4HeadDysphagia1142760181036705F18HeadDisfiguring lesion11381104521098506F12LegPain1123016381426729137F7HeadDisfiguring lesion19863127874158M3ParavertebralLocalization*1120087881118(+)29M13HeadDisfiguring lesion8100743225594910M11AbdominalLocalization**12045231821992211M6ThoraxPain11682632615535812F19HeadPain1086326153084213M13ParavertebralPain16657156315871214M16ParavertebralPain166531269941415F12HeadDisfiguring lesion30651436363216M10ThoraxPain15625046282817F14AbdominalPain1596104142548(+)3618F7ThoraxLocalization*184949128602 (+)15919M17ThoraxLocalization***114011601531335420M8PelvicPain1147110127708623721M9ParavertebralPain14812481882802422M11HeadLocalization***138741353731523F36LegPain131813010031823*Spinal cord compression, ** Bladder/urethral compression, *** Vascular/nervous compression Baseline evaluation of the patients: PN evaluated and correlated issue, interval between the pre-treatment MRI and the last one (days), length of the treatment (days), PN volume (cm3) before the treatment and at latest MRI evaluation, and volume change expressed as percentage; in the latter, (+) indicates a volumetric increase *Spinal cord compression, ** Bladder/urethral compression, *** Vascular/nervous compression Table 3Summary of volumetric changes in PNs during follow upTumor regression13 (57%)Tumor stabilization7 (30%)Tumor progression3 (13%) Volumetric reductionMedian23%Minimum4%Maximum70% Summary of volumetric changes in PNs during follow up All the adverse events that occurred during the follow-up have been reported in Table 4. Table 4Adverse events occurred during the treatment with selumetinib and their prevalence, expressed as percentageHigh CPK96%Paronychia91%Abdominal symptoms65%Acne65%Eczema and xerosis57%High ALT48%High AST44%Asthenia35%Peripheral edema26%Aftous lesions26%Irritability22%Hair loss22%Muscle cramps13%Lightened hair13%Pustules13%Itchiness9%Cataracts4% Adverse events occurred during the treatment with selumetinib and their prevalence, expressed as percentage The most frequent adverse event was CPK elevation, which happened in all but one patient (96%). The highest value identified was 1126 U/L (normal range 60–190 U/L), while the mean value was 390 U/L. In all the subjects of our cohort, CPK raised in the first months of treatment, and later returned within range in two years maximum. Interestingly, CPK values did not correlate with the appearance of muscular symptoms, neither in the patient with its highest peak. ALT and AST also increased during the follow-up period, respectively in 48% and 44% of the patients. However, their raise was significantly lower than CPK (maximum ALT value 78 U/L, maximum AST value 79 U/L). No patients developed hepatomegaly or hepatic complications during the treatment. Paronychia was extremely frequent among our cohort (91% of the patients). In 8 subjects, despite topical and oral antibiotics, a surgical approach was necessary, and one patient required it 4 times. Abdominal symptoms (diarrhea and abdominal cramps) happened almost exclusively in the first weeks of treatment in 65% of the patients and were treated with loperamide. Only one subject still complains of transient episodes of diarrhea. Dermatological side effects were particularly frequent. While adolescents and adults tend to develop acne, younger patients more likely develop eczema and xerosis. Among acne’s topical treatments, a combination of clindamycin and tretinoin was the most frequently used. On the other hand, lymecycline was the most prescribed oral antibiotic. No patient developed severe acne requiring oral isotretinoin. In the single patient that developed cataract (one year after starting the treatment), no surgical intervention was required since it remained minimal and stable over time. Two patients had a brief drug dose reduction, respectively due to diarrhea and asymptomatic CPK raise. After a month in which they took 20 mg/sqm twice a day, they returned to their standard dose. Only one patient permanently stopped the treatment with selumetinib, and not because of adverse events. In fact, after reaching adulthood, since her PN appeared stably reduced after two years of treatment, her family decided to try to stop the treatment, which we agreed upon. Nonetheless, the radiological and clinical follow-up was continued, and, in the following 9 months, the MRI scan confirmed the stability of the lesion and no new adverse events happened. Finally, it is important to point out that one patient who is now nine years-old suffered an ischemic stroke at the age of 6, one year after he began the treatment with selumetinib. The child already had a similar episode when he was 3-year-old, for which he was already taking aspirin as prophylaxis. Following the second episode, low molecular weight heparin prophylaxis was started as well, and selumetinib was temporarily stopped. However, after a discussion with the patient’s family, given the notable clinical benefits caused by selumetinib, the drug was restarted. Subsequently, a cerebral angiographic MRI showed the presence of Moya Moya disease, a rare vasculopathy associated with NF1 which causes terminal stenosis of the internal carotid artery, the anterior cerebral artery, and the middle cerebral artery. Overtime, the presence of multiple stenosis induces the development of new vessels of small caliber and insufficient flow: thus, it is likely that the stroke was caused by the Moya Moya disease itself, and not by the treatment with selumetinib. The patient underwent surgical treatment for this condition, and did not present similar episodes later, while continuing selumetinib. In terms of CTCAE grading scale, all the adverse events fall into grade 1 and 2, except for the upon described vascular event, which can be classified as grade 4. Discussion In this prospective case series, selumetinib appears to be useful in reducing PNs in patients affected by NF1. To better understand the efficacy of selumetinib in the treatment of PNs, we compared our data with those reported in the current medical literature [12–19]. Since only eleven articles came out, we eventually broaden the research including other molecules that are currently studied or under investigation, and seven other articles were found [20–26]. Their name and mechanism of action is reported in Table 5, while their effect in terms of regression of the PNs (if described precisely in the paper) is reported in Fig. 1. Currently, selumetinib seems the most promising drug in reducing the size of PNs. Table 5Current treatments available or under investigation for plexiform neurofibromasDrugMechanism of actionSelumetinibMEK1/2 inhibitorTrametinibMEK1/2 inhibitorMirdametinibMEK1/2 inhibitorBinimetinibMEK1/2 inhibitorCabozantinibMultiple tyrosine kinases and growth factors inhibitorAbemaciclibCyclin-dependent kinase (CDK4/6) inhibitorPirfenidoneAntifibrotic agent Current treatments available or under investigation for plexiform neurofibromas Percentage of partial response (reduction of more than 20% of the mass) recorded in patients treated for plexiform neurofibromas from different studies (with at least ten or more patients recruited). The black bars indicate the studies involving selumetinib, the grey ones the other drugs (McCowage: trametinib; Fisher: cabozantinib; Weiss: mirdametinib) Unfortunately, in our population we could not identify specific elements that were able to predict the clinical response to the treatment with selumetinib: in fact, neither the duration of the treatment, nor the age of the patient, nor the position of the tumor related to the mass reduction. In the past, Gross et al. hypothesized that an earlier treatment could be associated with a better outcome, suggesting that younger children could have an overall better response. Unfortunately, this assumption has not been confirmed by the studies currently available, but surely will be further investigated in the future, possibly on wider, transnational cohorts. As for the safety profile, this study confirms that selumetinib is a well-tolerated drug. No patient had to stop the treatment due to severe adverse events. In fact, adverse events were all minor or moderate. In comparison with the current literature, in our cohort we found a higher incidence of acne and paronychia. As for acne, no patient had to start isotretinoin. To lower the risk of paronychia, a great effort was made to sensitize patients and family on feet and periungual hygiene. Regarding the ischemic stroke that happened in the patient with Moya Moya syndrome, this event is more likely temporally associated with selumetinib rather than causatively, especially considering the history of the patient. However, due to severity of this manifestation, careful vascular surveillance should be considered for NF1 patients receiving long-term therapy. One of the key elements of this case series is the follow-up length which, to our knowledge, is the second longest in the literature. In fact, some our patients have been treated for more than six years. These data highlight that selumetinib action is substantially long lasting and reinforces the drug’s good safety profile. Another relevant element that we would like to underline, since it was never described in the past, is the fact that, during the follow-up period, PNs changed not only in size, but also in consistency and radiological appearance. Even since the very first months of treatment, patients often reported that they perceived their PNs as less tens and painful, before the appearance of a substantial radiological shrinkage. This perception was in fact associated with radiological changes in PNs signal intensity, especially in STIR sequences: while PNs appeared very well-defined in the pre-treatment MRI, their intensity gradually reduced during follow-up (Fig. 2). This radiological change was detected in all the patients who had tumor regression and stabilization, potentially accounting for the clinical benefit (especially pain reduction) that some patients descriptively reported despite the tumor stability. Change of intensity a plexiform neurofibroma of the calf: (a) before the treatment; (b) after 8 months of treatment This study has also various limits. The first one is that, for every patient, only a single PN was monitored throughout time. Therefore, we are not able to say with certainty whether some PNs respond to the treatment better than others, based on their location. Furthermore, inter-observer ed intra-observer variability among MRI evaluations was not assessed, nor were the evaluations blinded. The radiological change of the PNs (intensity reduction) and clinical effect on the patients is descriptive and will require further studies and clinical associations. Another limit is the follow-up time which, despite being remarkably long, is extremely variable among our cohort. The main concern is towards the patients who have a short follow-up (less than 2 years): based on the patterns that we identified in our population, their radiological response can be prolonged, but at the same time there’s a risk of losing the positive effect of the treatment and switch to a growing pattern. Furthermore, only one patient discontinued the treatment and was monitored for the following months, while, for the remaining patients, on-treatment time and follow-up period largely overlap. Thus, the risk of tumor regrowth or rebound and the development of long-term adverse events is not adequately assessed. The study also lacks quality-of-life data and pain scores, since questionnaires related to these areas were not collected before the beginning of the treatment in many patients of our cohort, and thus a comparison with post-treatment data would not be possible. Moreover, not all our cohort have a known pathogenic NF1 variant, and therefore it was not possible to evaluate a possible correlation between the patients’ genotype and clinical response. This occurred because multiple patients came to our attention from other clinical centers, and thus the results of the genetic investigation sometimes were not available. Conclusions In this prospective case series selumetinib appeared to be a useful and safe drug for the treatment of PNs. Moreover, in the current medical literature, selumetinib seems the most promising option among the various pharmacological options investigated. Nonetheless, despite these excellent premises, many questions remain unanswered about the treatment of NF1 and will need to be investigated in the future. The most relevant is the pharmacological approach that should be taken when patients stop responding or do not respond at all to selumetinib. It can be hypothesized that drugs with different actions, such a cabozantinib or ademaciclib, may have a role in this scenario. Another one is the intersection between selumetinib and surgery, and thus the possibility of using selumetinib as a neoadjuvant treatment, ensuring a volumetric reduction that may allow a surgical approach, or even as an adjuvant treatment, avoiding a potential regrowth of a plexiform neurofibroma after its resection. Finally, another element that should be investigated is the possible association of selumetinib with other drugs (such as mTOR inhibitors) to obtain a greater volumetric reduction, generating a larger block in PNs growth pathway. Hopefully, by addressing these issues, NF1 will finally embrace a precision medicine approach. Acknowledgements None to declare. Abbreviations Italian Medicines Agency alanine aminotransferase aspartate aminotransferase creatin phosphokinase electrocardiogram European Medicines Agency extracellular-signal-regulated kinase Food and Drud Administration mitogen-activated protein kinase malignant peripheral nerve sheath tumor magnetic resonance imaging neurofibromatosis type 1 plexiform neurofibroma rat-sarcoma rapidly accelerated fibrosarcoma short TI inversion recovery Authors’ contributions All authors contributed to the study conception and design. Data collection and analysis were performed by F.B., L.M. and A.A. Images were chosen by F.B. Tables were created by F.B. and L.M. The manuscript was written by F.B. and L.M. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Funding This work was supported by the Italian Ministry of Health, through the contribution given to the Institute for Maternal and Child Health IRCCS Burlo Garofolo, Trieste – Italy. Data availability All data generated or analyzed during this study are included in this article. Declarations Ethics approval and consent to participate All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Written informed consent was obtained from all individual participants included in the study or, in case of individuals younger than the age of 18, from their parents. Consent for publication Written informed consent for publication of clinical details and/or clinical images was obtained from the patients or their parents. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement All data generated or analyzed during this study are included in this article.