HEPZATO can be delivered only at a REMS-certified center. A patient reaches one through four sets of hands. Between the third and the fourth, no one is watching.
Interactive concept. Seeded, de-identified demonstration data. No patient information.
The device is approved and selling. Growth from here is a distribution problem: getting appropriate patients to a small, fixed set of certified sites.
Every number points at one lever. Each center sits at the end of a referral catchment: the practices that route patients to it. That catchment is where the next procedure comes from, and the part of the business no filing describes.
Uveal melanoma is rare, and the route from diagnosis to a certified center runs through several specialties over years. An ophthalmologist finds the tumor. An ocular oncologist treats the primary. Years later, a medical oncologist watching the liver finds the metastasis. Only then does a certified center enter, and only if someone makes the referral.
Surveillance is split by design, and the center sees the patient only once a referral exists. The pathway is real, but it is nobody's job. The systemic alternative does not close it: tebentafusp is limited to HLA-A*02:01–positive patients, roughly half.Approval For the rest, and for liver-dominant disease, the certified-center pathway is the access route.
new uveal melanoma cases a year in North America, about 5 per million. The most common adult eye cancer, and still rare.
SEERdevelop metastases, on average about three years after diagnosis. Surveillance has to hold across that gap.
SEER Consensusof metastatic cases have the liver as first or dominant site. Exactly the population HEPZATO is built for.
Cleveland ClinicThe eligible population is defined. The route to it is not.
No public source records how many reach a certified center, or where the rest stop. The loss is invisible because the pathway is not instrumented.
Each is a separate problem with a separate fix. From the field, all four look identical: a center that is not treating this quarter.
The window from treating the primary to finding a liver metastasis averages about three years, split across two specialties. A missed interval announces itself to no one.
A certified center is one option among several, and the oncologist making the call may have no certified site in view at all.
Thirty-one centers serve the whole country. Distance, scheduling, and coverage sit between a referral and a first visit, and delay to definitive treatment tracks with worse outcomes.Treatment delay
Once liver involvement passes the labeled threshold, or disease spreads beyond the liver, the pathway closes. Arrival is not eligibility.
Delcath already treats this as real. It launched a Healthcare Setting Locator in February 2024 and runs HEPZATO KIT Access 360, with a clinical navigator and copay support.Locator
Both exist because reaching the center is known to be hard. Neither shows which seam a given center is losing patients at, or which action a territory should take next.
Procedure volume per center, after the fact. A slowing center looks the same whether its referrers went quiet, its patients never arrived, or they arrived too late.
The position of the loss. Without it, every response is the same: another visit to the center that is already certified.
Phase 1 does not track patients. It instruments the pathway around each certified center at the account level, so the field can see where referrals stall and act on the specific stall.
For each of the nine territories, it maps the referral catchment of every center: which practices feed it, how recently, and where the flow thins. The unit is the referring practice and the center, never the patient. An intelligence and workflow layer, not a CRM, a REMS platform, or a patient hub.
A working concept of the Phase 1 board and territory funnel, on seeded, de-identified data.
Phase 2, following the patient, is the expansion: earned, not assumed. Patient-level tracking is where no patient falls through the cracks becomes literal, and it carries the consent, privacy, and security obligations Phase 1 does not. The order matters. Prove a non-PHI view of the pathway moves procedures first; then take on the consent architecture and PHI-grade security that patient-level tracking requires.
This brief shows the referral pathway can be mapped from evidence, and that the mapping is worth instrumenting. Every figure links to its record.
The full build is a paid engagement with Oddmind. This demonstration runs on hand-verified public data and carries no patient information.
Concept demonstration by Oddmind. All figures compiled from public sources in August 2026 and labeled where approximate. Not affiliated with or endorsed by Delcath Systems. HEPZATO KIT and Access 360 are Delcath's own program names, referenced as public fact.