Interview: Marlana Orloff | Professor, Department of Medical Oncology, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, USA
Citation: EMJ Oncol. 2026;14[1]. https://doi.org/10.33590/emjoncol/69C0L6C5
Uveal melanoma behaves very differently from cutaneous melanoma, particularly in where and how it spreads. What do we still not fully understand about the biology of this disease?
When people hear the word melanoma, most will think of skin or cutaneous melanoma. Though this is the most common type, there are many subtypes of melanoma, including uveal melanoma. Uveal melanoma starts in the melanocytes of the uveal tract in the eye, which includes the choroid, ciliary body, or iris. Uveal melanoma is distinctly different from cutaneous melanoma in its aetiology, mutations that drive it, pattern of spread, and how we treat it. Unlike cutaneous melanoma, where a majority are thought to be UV induced, uveal melanoma is not caused by the sun. The aetiology of uveal melanoma remains unknown. In cutaneous melanoma, we can often see BRAF or NRAS mutations, whereas a GNAQ or GNA11 driver mutation can be found in over 9% of uveal melanomas. One of the most important differences is uveal melanoma’s unique pattern of spread. In cutaneous melanoma, it can often initially spread via lymph node channels; however, uveal melanoma exclusively travels initially via the bloodstream and presents as liver metastases as first sight of metastatic disease in over 90% of patients. Given these varied patterns of spread, monitoring in uveal melanoma can be different from cutaneous melanoma. Lastly, given additional biologic differences, notably uveal melanoma’s lack of immunogenicity, the treatments of uveal melanoma are also different.
The liver is a common site of metastasis in uveal melanoma. Why does this cancer have such a strong predilection for the liver, and how does that influence the way you approach treatment?
Some of it has to do with the initial mechanism by which the cells can escape the primary tumour. They do so through the bloodstream, not really the lymph node channels, which is a big differentiator from cutaneous melanoma. There may be mechanisms that draw the cells to the liver and we are still trying to better understand that. Given the predominance of liver metastases in uveal melanoma, and the fact that liver involvement is often what can be life threatening, we can approach treatment by targeting the liver directly or treating systemically.
There is rationale for treating it systemically because we know the disease is inherently systemic. But again, because liver control is necessary, we also incorporate liver-directed therapy into many patients’ treatment course. There is a variety of liver directed options that can improve outcomes.
Tebentafusp was a major development in the treatment of metastatic uveal melanoma. As one of the trial investigators, could you explain how this treatment works and what makes its approach different from conventional immunotherapy, for those who may not be familiar with it?
When we think about conventional immunotherapy, we think immune checkpoint inhibitors like programmed cell death protein 1 (PD-1) and cytotoxic T lymphocyte-associated protein 4 (CTLA-4). Generally, they work by taking the brakes off the immune system and allowing the immune system to target the cancer. The immune system, though, because the brakes are taken off in a nonspecific way, can go after itself. Therefore, a lot of the side effects and risks are autoimmune diseases.
Standard immunotherapy, like immune checkpoint inhibitors, historically haven’t worked well in uveal melanoma. They’ve obviously changed the way in which we treat cutaneous melanoma, but effectiveness in uveal melanoma has been modest at best. There are a number of reasons we think immune checkpoint inhibitors may not work as well. Some of these reasons include uveal melanoma having less mutations, looking less foreign to the immune system, and also the metastatic dominance in the liver.
Tebentafusp is an engineered T cell engager/redirector that acts as a high affinity magnet that, on one end, binds any T cell glycoprotein 100 (GP100), which can be expressed on melanocytes or melanoma. Basically, this magnet drags the T cell over to the melanoma and forces it to engage. Because tebentafusp is linking someone’s own immune system to GP100 as a the target it does through human leukocyte antigen (HLA) presentation and therefore the drug is (HLA)-restricted. The drug was only designed to interact with patients who are HLA-A*02:01 positive, so we test patients’ HLA and determine whether or not they’re eligible for tebentafusp.
With longer-term data now available, what have the 5-year outcomes taught us about the durability of tebentafusp’s benefit?
We knew early on that tebentafusp doesn’t necessarily shrink tumours in a lot of patients. The response rates are modest. However, what we really learned early on is that even patients with stable disease can have a prolonged survival benefit. We have patients who have now been on the drug for 6 or 7 years. Patients who were part of the original clinical trial many years ago are now still on the drug. There’s definitely a subset of patients who with just a weekly tebentafusp infusion can have their disease controlled for a long time. It’s not making the cancer completely go away, but it’s certainly controlling the tumours for a long time.
The original overall survival data come from a 3-year follow-up and a 5-year follow-up. The 5-year follow-up is still suggesting that tebentafusp, over the investigator choice or control arm, is beneficial.
I think a concern is that from 3 years to 5 years of follow-up the tail of the curve did go down a little bit. So, while tebentafusp still looks better than the control group, which was a single agent immune checkpoint inhibitor, we were hoping to see a larger group of patients benefit for a long time. Patients are still benefiting, with some who are alive many years after starting tebentafusp. I think this fact supports its use as an option for our patients who are HLA-A*02:01 positive.
Your work has also examined hepatic delivery of melphalan (percutaneous hepatic perfusion or PHP) for unresectable metastatic uveal melanoma. What role do you see liver-directed therapy playing in the management of these patients?
There are a lot of options now, so whether it’s tebentafusp, PHP, or a number of other options including clinical trials, we look at every patient as an individual to make a decision about who would benefit from what.
Patients with liver dominant disease are certainly candidates for liver directed therapy, including PHP. There are patients who perhaps have too much disease outside their liver, where we would opt for systemic therapy versus PHP, or patients with >50% of their liver involved who would not be candidates for PHP. There have also been recent data with the CHOPIN trial,1 which combines PHP with the immune checkpoint inhibitors ipilimumab and nivolumab. Ipilimumab and nivolumab by themselves don’t work very well, but it turns out when you combine them with PHP, the outcomes are better than if one were to give each alone.
The CHOPIN approach is fairly standard practice now if patients don’t otherwise have a contraindication.
You recently contributed to a Delphi consensus on surveillance for metastatic uveal melanoma. How should surveillance evolve, and what would an ideal monitoring strategy look like for patients following treatment for primary uveal melanoma?
The Delphi consensus was a great effort to align surveillance practices. What we’re talking about is surveillance in the adjuvant setting; patients who’ve had their primary eye treated and you’re monitoring closely for metastatic disease.
I would say I would not be surprised that even with the Delphi consensus, there are still some institution- and provider-dependent surveillance protocols.
I believe we all agree that surveillance should be risk modified, by categorising patients on their risk of recurrence and then developing a surveillance strategy for them. We tend to categorise patients as low-, intermediate-, or high-risk, and that’s based both on tumour size/stage in the eye and genetic risk profiling, whether it be chromosomal analysis or gene expression profiling. Most will agree that patients should have frequent imaging of the liver to monitor for metastatic disease, often twice yearly for low and intermediate risk patients to start, and every 3 months for high risk patients. Chest imaging is also recommended at intervals based on risk.
At Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, USA, you work within a multidisciplinary team caring for patients with complex melanoma. How does this collaborative approach shape the way you manage patients with uveal melanoma, particularly those with metastatic disease?
In the metastatic setting, at Thomas Jefferson University Hospital we have an amazing multidisciplinary team. It’s comprised of medical oncology, interventional radiology (IR), radiation oncology, surgical oncology, and our research team. We have one of the most robust multidisciplinary groups. We actually see patients hand-in-hand along with our IR colleagues, and talk to patients about liver-directed therapy options, or other loco-regional options such as ablation or radiation. There are now clinical trials that also involve IR injecting things intratumourally.
For me, having this multidisciplinary approach is extremely meaningful. It is beneficial to have a variety of minds coming together to problem solve these complex cases. I think for patients it’s extremely meaningful too, because they feel like they have a whole team caring for them. It’s not that they’re just getting sent to this specialist or that specialist; there’s a lot of group decision-making and monitoring throughout the patient journey.
We used to consider the need for a multidisciplinary approach to patients with metastatic disease. However, it is clear that a multidisciplinary approach from Day 1 of diagnosis is critical. There are now options for patients to enrol on neoadjuvant clinical trials, which considers systemic treatment before the eye is even treated in an effort to shrink the primary eye tumour and reduce risk of recurrence. This means they need to meet the medical oncologist to screen them for such treatments before their eye is even treated. So, now more than ever, there’s really close collaboration and decision-making with the ocular oncologists.
For patients diagnosed with metastatic uveal melanoma, what are the biggest challenges they face beyond the treatment itself, and where do you think we could do more to support them throughout their cancer journey?
I think one of the main issues patients continue to struggle with is access to care. As uveal melanoma is a rare melanoma subtype, often, patients will have to travel to an expert centre that’s offering some of these therapies. We do our best to get patients’ care close to home, but that is not always possible. Travelling for care or trials can be quite expensive and could be a barrier to patients getting the most appropriate care.
We work closely with patient advocacy organisations to try to get patients to the best base, figuring out ways to get them financial support so they can get to these places for care. Also, continued education and just letting patients know how to advocate for themselves after being diagnosed with a rare cancer is important. Since uveal melanoma is rare and management differs from cutaneous melanoma, it is important that patients are seeing a provider who knows how to approach it.
Patients should be aware of this and know to seek an opinion at an expert centre if their current provider is unfamiliar.
Even though we have some effective treatments, and certainly more clinical trials than we had 10 years ago, I think a really big issue for uveal melanoma patients is the lack of very effective therapies. We’re still hoping for treatments that will provide a deeper, more durable response, and even cure.
