Multidisciplinary Decision-Making and CNS-Penetrant Therapies Redefine Brain Metastases Management in Lung Cancer

The growing availability of central nervous system (CNS)–penetrant targeted therapies is reshaping the management of brain metastases in select patients with lung cancer, challenging the long-standing priority of up-front local therapy, according to Laura Alder, MD.

A recent publication by Alder et al suggested that patients with asymptomatic CNS disease should be afforded the opportunity to begin treatment with a tyrosine kinase inhibitor (TKI) before considering radiation, whereas those with large, symptomatic brain metastases or lesions in critical locations may be better suited for surgical resection and stereotactic radiosurgery (SRS).¹

“We have to look at what’s going on systemically and in the CNS. How big are the brain lesions? Are they symptomatic? What’s the location of these brain lesions? Are they in very sensitive areas of the brain where any growth could be catastrophic? What’s the associated edema? Is it single? Is it multiple? All these conversations require multidisciplinary involvement, because at the end of the day, we want to make sure that we’re giving the patients the very best outcomes, and that we limit toxicities and think long-term,” Alder said in an interview with OncLive®.

Alder is an assistant professor of medicine at Duke University School of Medicine and a member of Duke Cancer Institute in Durham, North Carolina.

In the interview, which was conducted during the Bridging the Gaps: Consensus on Brain Metastases meeting,² Alder discussed how treatment decisions in CNS-only disease are increasingly individualized, driven by tumor genomics, lesion characteristics, and evolving multidisciplinary input. She also highlighted ongoing gaps in defining oligoprogression, interpreting complex imaging findings, and integrating emerging tools such as cerebrospinal fluid sequencing into ongoing research.

OncLive: When should local therapy be prioritized over systemic therapy in CNS-only disease?

Alder: I recently published a publication with Dr Sandeep Patel and Dr Thomas Stinchcomb in the Journal of Thoracic Oncology. It’s a controversial piece, and some of the things that we discussed [illustrate that the answer] is very patient dependent. We’re looking at what drugs are available in lung cancer, because we have some CNS-penetrant drugs. These are often oral TKI therapies for EGFR, ALK, ROS1, and RET alterations, et cetera, that can achieve brain penetrance of upwards of 80% to 90%, so when we have such a therapy, we prefer to start that first, with one caveat.

We have to look at what’s going on systemically and in the CNS. How big are the brain lesions? Are they symptomatic? What’s the location of these brain lesions? Are they in very sensitive areas of the brain where any growth could be catastrophic? What’s the associated edema? Is it single? Is it multiple? All these conversations require multidisciplinary involvement, because at the end of the day, we want to make sure that we’re giving the patients the very best outcomes, and that we limit toxicities and think long term.

So, if we do start a systemic therapy first, our practice is to repeat a brain MRI, usually [within] 4 to 6 weeks, to ensure that we’re seeing some sort of response, because with a lot of TKIs, we do see [an] early response if they’re going to respond. Most of us feel that that time frame is enough to get a good idea [of whether a response is imminent], and if they’re not responding as we’d like, go back to the multidisciplinary board. Most likely, we’re going to proceed with some sort of local treatment, usually SRS.

Does the choice of systemic therapy affect how you approach these decisions?

Immuno-oncology drugs are less CNS penetrant, so we have a higher threshold to pull the trigger when we’re talking about local therapy, such as radiation or even surgery if it’s only one symptomatic lesion, because we really don’t want to miss the chance to tackle something when we first see it, and we don’t want to give [the lesion] the chance to grow. Again, [this is where] multidisciplinary discussions [come into play so we can] make sure we’re choosing the best outcome, making sure we do an MRI [scan] of the brain for every patient at initial diagnosis. Whenever patients are progressing systemically, I like to repeat a brain MRI [scan] because we want to ensure that we’re not missing anything. The earlier we catch these things, the more options we have available.

How many progressing sites define CNS oligoprogression that remains amenable to local therapy and continuation of systemic therapy?

We’ve talked a little bit about the gap in what the optimal surveillance is for brain MRI imaging, especially when starting systemic treatment first. There’s not a great definition for oligoprogressive disease in the brain and systemically. It really varies. Clinical trials have various definitions. Guidelines have various definitions. So I ask, “Can we do SRS instead of whole brain radiation?” As much as possible, we try and avoid whole brain radiation. With respect to SRS, many years ago, I believe [the answer] was 3 to 5 lesions. That number is changing rapidly, and it depends on the institution, the type of radiation, the fractionation, where the brain metastases are. But I’ve heard sometimes that 20 [lesions] can still be amenable to SRS. We need more consensus from multidisciplinary expertise that involves our radiation oncology [colleagues].

How should ambiguous CNS imaging findings be managed?

One, as much as possible, we always want the brain MRI with and without contrast [because] that’s more detailed than a CT scan. One of the things that we have trouble with is, let’s say the patient had radiation in the past. Is this radiation necrosis, or is this CNS progression? That’s especially complicated if only the brain is showing signs of changes and everything in the body is stable. So again, a multidisciplinary team can be great. We have a few new imaging techniques with brain MRIs, looking at diffusion, different types of modalities that we can talk to a radiologist about to see whether we can pinpoint and understand what this is without the need for a biopsy, because that’s the most definitive. We know tissue is always the gold standard, but are there things we can do along the way that can maybe prevent a more invasive tissue biopsy?

One of the things that we’re gaining more information about is the use of liquid biopsies, such as plasma, [to see whether] that can pick up some of the circulating tumor DNA from active CNS metastases. One of the things that’s gaining more attention is cerebrospinal fluid [CSF] genomic sequencing and how that can possibly be integrated to diagnose and then select optimal treatment, too.

What are some current best practices surrounding both approaches?

Every new diagnosis of lung cancer, regardless of the stage, needs sequencing, because we use that in every decision. We use that in the early-stage setting, [when we’re considering] neoadjuvant [therapy] before surgery. We have approved treatments in the EGFR space now before surgery, and then after surgery, we have them in the EGFR and ALK space as well. We know immunotherapy doesn’t work as well in these patients, and then the metastatic setting, genomic sequencing should be done as soon as possible at diagnosis. I usually like to send the liquid [biopsy] as soon as I meet the [patient], because that can come back quickly while we’re waiting for tissue. If that comes back with an answer, we have a great treatment that we can start quickly and get insurance approval and everything rolling.

The other time that I like to consider repeat genomic testing is at progression. Tissue is the gold standard, especially if you’re worried about EGFR transformation to small cell. But if the patient is being treated with a targeted therapy, we want to see if there are any resistance mechanisms to that targeted therapy. That’s when that genomic testing result will be huge. One of the difficulties when you have CNS-only progression, so either in the brain tissue itself, parenchymal or leptomeningeal disease, is that [the results] from data plasma are [approximately] 50/50, overall. That’s why some of the new research looking at CSF genomic analysis is promising, because that saves the patient from undergoing a biopsy. Sometimes even CSF biopsies are not always as specific and sensitive as we’d like, but the CSF can give us the best tools to make sure we treat that patient with the best next-line option too.

References

1. Alder L, Patel SP, Stinchcombe TE. Given TKIs with high CNS penetrance, systemic therapy should be administered before radiation therapy for CNS metastases. J Thorac Oncol. 2025;20(12):1757-1759. doi:10.1016/j.jtho.2025.09.006
2. OncLive® Bridging the Gaps: Consensus on Brain Metastases. OncLive. January 21, 2026. Accessed March 16, 2026. https://events.bizzabo.com/810556/home?widget=true