Scientists identify dual control mechanism for 'undruggable' MYC cancers

For two decades, Anindya Bagchi carried an unsolved puzzle from his own doctoral work. It concerned MYC, a gene that sits at the center of cancer's most aggressive machinery—a master switch that controls cell growth, division, and metabolism. When MYC goes wrong, which happens in more than half of all human cancers, cells spiral into uncontrolled multiplication. Tumors become treatment-resistant. Patients face grim odds. Yet MYC has remained stubbornly resistant to the drugs that work elsewhere in cancer biology. The protein lacks the binding pockets that conventional medicines target. Clinicians have had few direct ways to fight back.

This month, Bagchi and his team at Sanford Burnham Prebys published not one but two papers, back-to-back in Genes & Development, that finally crack open how MYC actually works—and more importantly, how to attack it sideways. The key lies not in MYC itself, but in a neighboring stretch of DNA on chromosome 8, a region called PVT1 that had long seemed like a passive bystander. Previous work in Bagchi's lab showed PVT1 was essential for MYC-driven tumors to grow, but nobody understood why. The new studies reveal it is far more than a neighbor. It is an active control hub, encoding two entirely novel proteins that either fuel or restrain MYC's cancer-driving power.

The first discovery emerged from examining how PVT1 generates a circular RNA that codes for a protein the researchers named Firefox. When cells contain Firefox, MYC becomes far more potent—it produces more of itself and drives the molecular cascade that transforms normal cells into cancer cells. When researchers depleted Firefox in laboratory models of MYC-driven cancers, tumor growth slowed significantly. Firefox, Bagchi explained, acts as a critical helper that MYC needs to fully transform a cell. Remove it, and MYC loses much of its destructive power.

The second discovery is equally striking. The researchers found that structural rearrangements at PVT1—moments when a DNA segment breaks away and attaches elsewhere—consistently delete a region that encodes a micropeptide they named Honeybadger. This micropeptide normally acts as a brake on cancer growth. It binds directly to KRAS, a key signaling protein implicated in 25 to 30 percent of all human cancers. By binding KRAS, Honeybadger dampens the RAS-MAPK signaling pathway under normal conditions. But when PVT1 rearrangements delete the Honeybadger-encoding region, that brake vanishes. Wild-type KRAS becomes hyperactive, which in turn stabilizes MYC protein and amplifies its output—even in tumors that lack KRAS mutations themselves.

What makes this discovery particularly powerful is the dual mechanism at work. A single structural alteration at PVT1 simultaneously accomplishes two things: it preserves Firefox, an oncogene that fuels MYC, and it eliminates Honeybadger, a tumor suppressor that normally restrains it. Both changes converge to amplify MYC activity. This explains why PVT1-rearranged cancers carry such a poor prognosis. It also opens a door that has been locked for decades. "You do not have to hit MYC directly to control MYC-driven cancers," Bagchi said. By targeting the molecules MYC depends on—Firefox and Honeybadger—researchers have identified entirely new therapeutic entry points for tumors that have long seemed beyond the reach of targeted drugs.

The implications ripple outward. PVT1 and its encoded proteins now represent a rich source of biomarkers that could help identify which patients will respond to new treatments. More immediately, they offer prototype drug targets. Bagchi's team plans to investigate how Firefox and Honeybadger behave across additional cancer types and to begin developing therapeutic strategies with collaborators. The goal is to translate these basic discoveries into first-in-class treatments for patients with MYC-driven tumors—a population that currently has very limited options. After two decades, the puzzle that haunted Bagchi's graduate work may finally be yielding its answer.