The cancer comes back, now resistant to the drugs that briefly held it at bay.
For decades, glioblastoma has defied medicine's best efforts, returning after treatment as something harder and more resistant than before. Now, researchers at UT Southwestern and the University of Alabama at Birmingham have identified a molecular conversation between two proteins — EGFR and MGMT — that may explain why chemotherapy eventually fails, and how, with careful timing, that failure might be prevented. The discovery does not yet reach patients, but in a disease where almost nothing has changed in a generation, a plausible mechanism is itself a form of hope.
- Glioblastoma kills roughly 250,000 people a year within months of diagnosis, and no treatment has meaningfully altered that reality in decades.
- The cancer's cruelest trick is adaptation: the standard chemotherapy drug TMZ actually trains tumors over time to produce more of the very protein — MGMT — that neutralizes it.
- Researchers discovered that blocking EGFR, a receptor frequently mutated in glioblastoma, also shuts down MGMT production, stripping tumors of their chemical armor and restoring sensitivity to chemotherapy.
- Timing proved to be everything — the EGFR inhibitor must be given a day before chemotherapy, not alongside it, a sequencing detail that may explain why earlier combined trials quietly failed.
- The findings are preclinical, but they reframe past failures as a solvable problem of sequence rather than an insurmountable wall of biology.
Glioblastoma offers its patients a brutal arithmetic: a quarter million diagnoses each year, median survival of twelve to eighteen months, and a five-year survival rate hovering between five and seven percent. Surgery, radiation, and chemotherapy buy time, but the cancer almost always returns — now resistant to the drugs that once slowed it. For that recurrence, medicine has had almost nothing to offer.
The standard chemotherapy, temozolomide, works by damaging cancer cell DNA. But glioblastoma carries a countermeasure: a protein called MGMT that repairs exactly that damage. Worse, TMZ itself causes tumors to produce more MGMT over time, effectively teaching the cancer to survive treatment.
Researchers at UT Southwestern Medical Center and the University of Alabama at Birmingham, led by Amyn Habib and Gao Guo, were studying EGFR — a receptor gene commonly mutated in glioblastoma — when they made an unexpected discovery. Blocking EGFR also suppressed MGMT production, in both lab-grown cells and mouse models. Introducing the EGFR inhibitor afatinib made resistant tumors vulnerable to TMZ again.
The critical variable was sequence. When afatinib was given one day before chemotherapy, the combination worked. When both drugs arrived simultaneously, it did not. MGMT had to be cleared first. This detail reframes a history of failed clinical trials: previous studies tested the drugs together, and tumor samples from those trials showed MGMT levels remained high. Samples from trials using only an EGFR inhibitor showed the protein significantly reduced — suggesting that chemotherapy delivered afterward might have succeeded.
The research remains preclinical and years from clinical application. But for a disease that has resisted meaningful progress for so long, identifying a plausible mechanism — and a reason why past attempts fell short — is a rare and significant step.
Glioblastoma kills most of its victims within eighteen months. The median survival stretches to perhaps two years if treatment works, but it almost never does for long. About a quarter million people worldwide receive this diagnosis each year, and for nearly all of them, the prognosis remains essentially unchanged from decades ago: surgery, radiation, chemotherapy, and then the cancer comes back, now resistant to the drugs that briefly held it at bay. There are no effective options for what comes next.
Researchers at UT Southwestern Medical Center and the University of Alabama at Birmingham have identified a potential way to break this cycle. In preclinical studies published in Science Translational Medicine, they found that a class of drugs called EGFR inhibitors can restore glioblastoma's sensitivity to chemotherapy by targeting a specific resistance mechanism. The work, led by Amyn Habib and Gao Guo, suggests a new treatment sequence that could eventually become standard care for this devastating cancer.
The standard chemotherapy for glioblastoma is temozolomide, or TMZ. It works by damaging cancer cell DNA, preventing division and triggering death. But the disease has a cruel pattern: patients who initially respond to TMZ almost always develop resistant tumors. The reason lies in a protein called MGMT, which repairs the very DNA damage that TMZ inflicts. Tumors that produce high levels of MGMT can shrug off the drug. Even worse, TMZ itself prompts cells to make more MGMT over time, essentially training the cancer to resist treatment.
Habib's lab has spent years studying the epidermal growth factor receptor, or EGFR, a protein produced by a gene frequently mutated in glioblastoma. While investigating which molecular pathways EGFR controls, the researchers made an unexpected discovery: blocking EGFR also shut down MGMT production. They saw this effect both in glioblastoma cells grown in the laboratory and in mouse models of the disease. When they added an EGFR inhibitor called afatinib, the tumors became sensitive to TMZ again—even tumors that had already developed resistance.
But there was a crucial detail. The timing mattered absolutely. When researchers gave afatinib one day before TMZ, the combination worked. When they gave both drugs at the same time, nothing happened. MGMT had to be suppressed first, before chemotherapy could do its job. This finding may explain why previous clinical trials testing TMZ and EGFR inhibitors together produced disappointing results. Habib's team examined tumor samples from one such failed trial and found high MGMT levels in the cancer cells—the drugs were fighting each other, not the disease. In contrast, samples from a trial testing only an EGFR inhibitor showed significantly reduced MGMT, suggesting that chemotherapy delivered afterward might finally work.
The five-year survival rate for glioblastoma remains between five and seven percent. If future clinical trials confirm that this sequencing strategy—EGFR inhibitor first, then chemotherapy—can restore drug sensitivity in patients, it could transform outcomes for a disease that has resisted meaningful progress for decades. The work is still preclinical, still years away from changing how patients are treated. But for a cancer with almost no good options, it represents something rare: a plausible path forward.
Notable Quotes
Glioblastoma is a devastating brain cancer with a dismal prognosis and no truly effective treatments. Our study could give new hope to the approximately 250,000 patients worldwide diagnosed with this disease each year.— Amyn Habib, M.D., UT Southwestern Medical Center
The Hearth Conversation Another angle on the story
Why does the timing matter so much? Why can't you just give both drugs together?
Because MGMT is the lock and EGFR inhibition is the key. You have to turn off the lock-making machinery first. If you give both drugs at once, the cancer cells are still churning out MGMT while the chemotherapy is trying to work. The repair protein wins.
So in those failed trials, they were essentially giving the cancer a shield and a sword at the same time?
Exactly. They were trying to hit the cancer with chemotherapy while the cancer was actively producing the protein that protects it from chemotherapy. The researchers looked back at those trial samples and found high MGMT levels. The strategy was right, but the execution was backwards.
How confident are they that this will work in actual patients?
This is preclinical work—lab cells and mice. The next step is clinical trials. But Habib thinks if those trials confirm the sequencing works, this could become the standard treatment. That's a big if, but the logic is sound.
What's the human stakes here?
Two hundred fifty thousand people diagnosed every year. Most dead within eighteen months. Five-year survival around five to seven percent. This isn't a rare disease. It's a common death sentence. Any real improvement matters enormously.