RTOG 9402: What the Brain Tumors Trial Found

A trial that missed its target and changed practice anyway

RTOG 9402 set out to answer one question: for a brain tumor called anaplastic oligodendroglioma, does adding chemotherapy to radiation help people live longer?

For the group as a whole, the answer was no. Median survival was 4.6 years with chemotherapy plus radiotherapy against 4.7 years with radiotherapy alone. The hazard ratio was 0.79 (95% CI 0.60 to 1.04, P=0.1). That is a failed primary endpoint.

And yet this trial reshaped how brain tumors are diagnosed. The reason is what turned up when the tumors were sorted by their genetics.

The tumor and the treatment

Oligodendrogliomas start in oligodendrocytes, the brain cells that wrap nerve fibers in insulating myelin. "Anaplastic" meant the cells looked disordered and fast-growing under a microscope.

291 eligible patients with anaplastic oligodendroglioma, or with a mixed tumor called anaplastic oligoastrocytoma, were randomly assigned:

• PCV chemotherapy plus radiotherapy: 148 patients. PCV is procarbazine, lomustine, and vincristine.
• Radiotherapy alone: 143 patients.

Results were published in the Journal of Clinical Oncology in 2013, after roughly 12 years of follow-up. Earlier reports had shown no benefit at all.

The genetic split

Some of these tumors are missing pieces of two chromosomes — the short arm of chromosome 1 and the long arm of chromosome 19. This is called 1p/19q codeletion.

Sorting patients by that marker separated the trial into two different stories.

With 1p/19q codeletion: median survival was 14.7 years with PCV plus radiotherapy against 7.3 years with radiotherapy alone. Hazard ratio 0.59 (95% CI 0.37 to 0.95, P=0.03). Roughly double.

Without the codeletion: 2.6 years against 2.7 years. Hazard ratio 0.85 (95% CI 0.58 to 1.23, P=0.39). No difference whatever.

The marker was also powerfully prognostic on its own. Among patients given both treatments, those with the codeletion lived a median of 14.7 years against 2.6 years for those without it. On radiotherapy alone the figures were 7.3 against 2.7 years. Both comparisons were highly significant.

Why this mattered beyond one trial

Two groups of patients, indistinguishable to the naked eye, had opposite answers to the same question. One group gained years from chemotherapy. The other gained nothing and absorbed the toxicity for free.

Molecular testing stopped being optional in brain tumor diagnosis. NCI's current guidance treats 1p/19q codeletion, along with IDH gene variants, as part of how these tumors are classified and how prognosis is judged. Our brain tumors page covers how that classification works today.

How a brain tumor announces itself

There is no screening test for brain tumors. Symptoms come from a growing mass pressing on tissue that has nowhere to expand into. See a clinician promptly for:

• A new headache pattern, particularly one worse in the morning, worse when lying flat, or worse on coughing or bending.
• A first-ever seizure at any age, which needs emergency assessment.
• Weakness, numbness, or clumsiness on one side of the body.
• Trouble finding words, or speech that has become slurred.
• Changed vision: blurring, double vision, or losing part of the visual field.
• Personality or memory changes that family members notice before you do.
• Persistent nausea or vomiting with any of the above.

Most headaches are not tumors. A headache that is genuinely new in character, particularly with any neurological change, is worth same-week attention.

What this trial cannot tell you

• The trial failed its primary endpoint. Overall survival across the whole group was not improved, at P=0.1.
• The codeletion finding came from what the authors themselves describe as an unplanned analysis. That is a substantially weaker grade of evidence than a prespecified one.
• Codeletion status was known only for tumors with usable tissue, so the subgroup result does not rest on all 291 patients.
• PCV is hard to tolerate. No randomized trial has compared it head to head against temozolomide, which is easier on patients, in this setting.
• It took about 12 years for the difference to appear. A trial reporting at 5 years would have concluded the opposite.
• Tumor classification has been rewritten since 2013, so the diagnostic labels used here do not map cleanly onto today's. Trial phases sets out what a phase 3 result can support.

Sources

This article was written from the sources below, which were checked on the source-check date shown above.

• Cairncross G and colleagues, "Phase III trial of chemoradiotherapy for anaplastic oligodendroglioma: long-term results of RTOG 9402," Journal of Clinical Oncology, 2013 (PMID 23071247): https://pubmed.ncbi.nlm.nih.gov/23071247/
• National Cancer Institute, Central Nervous System Tumors Treatment (PDQ), health professional version: https://www.cancer.gov/types/brain/hp/adult-brain-treatment-pdq
• National Cancer Institute, Brain and Spinal Cord Tumors: https://www.cancer.gov/types/brain

How this page was made

An AI-assisted editorial system helped prepare this page. No named medical reviewer has reviewed it unless one is listed.

The National Cancer Information Foundation publishes Cancer Explained. It is not a diagnostic service, does not recommend treatments, and is not for emergencies.

See an error, old source, or unclear wording? Tell us.