Back Back
NF1 Challenges
Real-World Challenges With MEK Inhibitors for NF1-Related Plexiform Neurofibromas

Released: August 02, 2021

Expiration: August 01, 2022

Scott Plotkin
Scott Plotkin, MD, PhD
Activity Information

Activity

Progress
1
Course Completed
Continue

In the decades since the NF1 gene was first cloned in 1990, a great deal has been learned about neurofibromin, the protein encoded by NF1. Loss of neurofibromin function leads to activation of the RAS pathway and, most commonly, activation of the MAP kinase pathway, including Raf, MEK, and ERK. Individuals with germline pathogenic variants in NF1 have neurofibromatosis type 1 (NF1), predisposing them to develop plexiform neurofibromas, gliomas, other tumors, and learning disabilities. 

In this commentary, I discuss several important issues when using MEK inhibitors to manage NF1.

Choosing a MEK Inhibitor
Selumetinib is currently FDA approved for pediatric patients with NF1 and symptomatic, inoperable plexiform neurofibromas. Other MEK inhibitors are available through clinical trials or off‑label. However, MEK inhibitors have not been compared head‑to‑head in trials, and there are important differences in the clinical trial design that make comparisons difficult. For example, trials NF108 (binimetinib) and NF106 (mirdametinib) run by the Department of Defense Neurofibromatosis Clinical Trials Consortium were designed for a rapid readout—meaning that if the patient did not achieve at least 15% tumor shrinkage in 8 months, they were removed from the study, whereas those with at least 15% shrinkage continued for up to 24 months. By contrast, the SPRINT trial evaluating selumetinib did not require a minimum amount of tumor shrinkage and permitted participants to stay on study as long as they did not have progressive disease.

Assessing Efficacy in the Real World
Precisely measuring tumor response for plexiform neurofibroma remains an ongoing challenge in community practice. One option is to perform a linear measurement using Response Evaluation Criteria in Solid Tumors (RECIST) criteria or a 2-dimensional measurement, but these approaches may not accurately measure irregular or diffuse lesions. One solution is to perform volumetric MRI—a computer‑assisted technique in which a human operator outlines the tumor slice by slice on MRI scan to calculate its volume. This invaluable yet labor-intensive technique allows detection of small but important changes in tumor size, allowing healthcare professionals to detect when tumors stop responding to treatment. It also enables more rapid determination of response; an analysis performed by the National Cancer Institute reported that compared to volumetric analysis, standard 2-dimensional measurements would take 1-2 years longer to detect treatment response.

The practical challenge is that volumetric MRI is not widely available in the community. Most clinicians do not send MRIs for volumetric analysis and instead assess response with either linear or 2-dimensional analyses. When precise volumetric analysis is not available, clinicians often assess clinical benefit based on symptoms: Does pain decrease? Does function improve? In these scenarios, clinicians de-emphasize MRI scans. For those patients who do not experience benefit in quality-of-life or MRI scans, toxicity frequently leads to discontinuing the drug.

What about treating asymptomatic patients with known plexiform neurofibromas? In this situation, I determine whether the tumor is growing by performing MRI scans every 1-3 years (depending on the clinical scenario). If the tumor is asymptomatic and not growing, I would not treat with MEK inhibitor. If the tumor is asymptomatic but growing, I then consider the risk it poses to the patient. For example, if it is growing near the airway or other vital structure, I would likely present the case at our tumor board and consider treating. Otherwise, I might choose to just monitor an asymptomatic, growing, yet low-risk tumor because longitudinal data using volumetric MRI indicate some tumors spontaneously regress. I want to make the point that there is no substitute for good clinical judgment, and we follow these patients closely because sometimes the tumor might grow for a period and then not grow or decrease. If we fall into the trap of treating every tiny tumor that is growing but asymptomatic, we often will give patients more toxicity from chemotherapy or neurologic disability from surgery.

Optimal Duration of Therapy
We do not yet know the optimal treatment duration with selumetinib. The median time to response in the SPRINT trial was about 12 months and to best response was 2 years. In the real world, coaching patients on expectations for response and side effects is important. In addition, the published SPRINT trial data and subsequent experience with real-world patients on selumetinib demonstrated that, for many individuals with good response, a treatment hold can lead to tumor regrowth. In some of these individuals, retreatment with selumetinib can lead to tumor shrinkage. Thus, long‑term treatment may be warranted. We need studies on long-term outcomes as well as on alternative dosing regimens that might improve long‑term tolerability—perhaps intermittent treatment as performed with mirdametinib (3 weeks on, 1 week off schedule). 

Activity and Tolerability of MEK Inhibitors in Adults and Children
Another important quality-of-life issue with MEK inhibitors is their dosing schedules, which can range from once a day (eg, trametinib) to twice a day (eg, selumetinib). MEK inhibitors require an empty stomach for 2 hours before the dose and 1 hour after. This is particularly challenging for a school-age child, because if they get their first dose at 7 a.m. after waking up, then they either have to eat dinner and be done by 5 p.m. or not start eating until 8 p.m.—a difficult scenario that affects quality of family life. 

There is another important difference with MEK inhibitors in adults vs children. The side effects with treatment vary with age—children rarely experience skin toxicity while teenagers and adults frequently develop a rash. Although the efficacy appears to be comparable between the populations, toxicity poses more of a challenge in adults and leads to more patients not receiving full doses due to holds or discontinuations. There is a pressing need to develop more tolerable agents and better supportive care for key adverse events such as acneiform rash.

An aggressive preventive approach for acneiform rash is needed because the rash is easier to prevent than to treat. The best practice for the management of acneiform rash is prevention and many clinicians start the patient on clindamycin gel or lotion before initiating selumetinib. Because clindamycin can cause skin irritation, a slow introduction at once a day for a week and then increase to twice a day is helpful. Depending on the patient’s age and their skin health before starting selumetinib, we sometimes also initiate doxycycline or minocycline. 

MEK Inhibitors for Other NF1-Related Tumors
MEK inhibitors are under investigation in other tumor types driven by NF1 loss. For example, the Pediatric Brain Tumor Consortium is conducting a phase II trial of selumetinib in pediatric patients with low‑grade gliomas, including a stratum of patients with NF1-associated glioma. When looking at best responses in this stratum, I was amazed by how 40% of patients achieved a partial response and 60% had stable disease; remarkably, only 1 case of progressive disease was observed during treatment. The progression‑free survival (PFS) was remarkably good with a 2‑year PFS rate exceeding 90%.

Selumetinib was fairly well tolerated, with 16 participants completing all 26 courses. The safety profile was comparable to that observed in plexiform neurofibroma. These are encouraging data, and the Children’s Oncology Group is recruiting for phase III trials comparing selumetinib vs chemotherapy in patients with low-grade gliomas that are either NF1-associated (NCT03871257) or sporadic (NCT041660409).

Selumetinib is also under investigation in an ongoing phase II trial run by the University of Alabama at Birmingham and the National Cancer Institute for adults with cutaneous neurofibroma (NCT02839720). The primary endpoint is shrinkage of the cutaneous neurofibromas. We hope to see results in the next several years to better understand MEK inhibitor activity against this very disfiguring manifestation of NF1.

Looking to the future, there is a planned trial investigating combining a MEK inhibitor with cabozantinib, and there is some interesting research into the biology of tumor progression and subsequent treatments.

Your Thoughts?
What challenges have you encountered when managing NF1 with MEK inhibitors? Please answer the polling question and join the conversation by posting a comment to the discussion section.

Continue

Poll

1.
What would you like to learn more about related to the care of patients with NF1 and associated malignancies?
Submit