Glioblastoma's Dismal Standard of Care: The Stupp Protocol

By early April I was at home, doing my best to forget the worst two weeks of my life. I'd been diagnosed with a tennis ball-sized brain tumor, which had been surgically removed. Both imaging and pathology were consistent with glioblastoma, a fatal, rare, incurable, and ferocious form of cancer. But even with that news harshing the vibe, I was enjoying my new freedom — from sensors, from bandages, from needles, from embarrassing cognitive assessments, from bathroom surveillance, from bizarre late-night procedures, and from institutional odors.

All the same, I knew this wasn't the end but rather a transition. To what I had no idea. The talk of my neuro-oncologist didn't square with the surgery team's announcement that they had "got it all." As if to drive home the message that this wasn't over by a long shot, somebody had peppered my calendar with new appointments, the purpose of which I could only guess. Most of the doctors' names were unknown to me. Oddly, I'd told nobody that I was even interested in further treatment.

It would take a few weeks for me to piece together what was happening: this was the opening act of the glioblastoma "standard of care." The term "standard of care" is used throughout medicine to mean something very specific. Donna Vanderpool defines it like this:

… The standard of care is the benchmark that determines whether professional obligations to patients have been met. Failure to meet the standard of care is negligence, which can carry significant consequences for clinicians. … The standard of care is a legal term, not a medical term. Basically, it refers to the degree of care a prudent and reasonable person would exercise under the circumstances. …

I'd later learn that the specific standard of care for glioblastoma was something called the Stupp Protocol (aka "Stupp Regimen").

The Stupp Protocol takes its name from Roger Stupp, first author on a landmark 2005 paper reporting the results of a successful Phase 3 clinical trial. It consisted of two phases: (1) radiation therapy with concomitant chemotherapy; and (2) adjuvant chemotherapy. In the first phase, patients would receive radiation therapy while taking an oral chemotherapeutic drug. In the adjuvant phase, patients would take the chemotherapeutic drug only.

Like most clinical work, Stupp's trial built on the work of others. By the late 1970s, clinical trials had demonstrated the effectiveness of drugs like carmustine for treating gliomas, both in concomitant and adjuvant modes with radiation therapy. Although they use different chemotherapeutic agents, the overall structure of these early protocols resembles the one that Stupp's team reported on in 2005.

What all of these efforts stretching back over 40 years or more have in common is the use of a brain-penetrating alkylating agent as the chemotherapeutic agent. The idea is simple. In normal adult brain tissue, cell division is unusual. But in gliomas, cell division is common. Damaging DNA with drugs like carmustine and temozolamide disrupts cell division chemically, especially through the attachment of carbon atoms to certain oxygen atoms in DNA. Having sustained this damage, rapidly-dividing cells such as those found in gliomas can't reproduce as quickly, which ultimately leads to their death.

Healthy brain cells that don't divide as rapidly suffer relatively less damage. Even so, the fact that some damage is caused in the brain and elsewhere nevertheless explains some of the side effects of alkylating drugs used to treat glioblastomas. One of the most serious of these is lymphocytopenia, a drop in white blood cell count. As a result, the Stupp protocol requires patients take an antibiotic during the concurrent phase and encourages it during the adjuvant phase.

The close similarity in how these drugs behave raises the question: Why temozolomide? In 2005 there were other alkylating drugs that Stupp's team could have used.

An important advantage of temozolomide is that it crosses the blood-brain barrier (BBB). One of the BBB's main duties is to protect the brain from small molecules like temozolomide and the damage they can cause. Engineering drugs capable of crossing the BBB is difficult, but quite achievable. (During my career as a medicinal chemist, I did it many times for pre-clinical projects.) By 2005, Several brain-penetrating alkylating agents would have been available for Stupp to choose from.

Another factor could have been financial. The acknowledgements section of the 2005 paper notes that the research was supported by "… an unrestricted educational grant from Schering-Plough, Kenilworth, J.J., which also provided the study drugs," and that "… Drs. Stupp, Mason, van den Bent, Brandes, Cairncross, and Mirimanoff report having received consulting and lecture fees from Schering-Plough." I wouldn't read any kind of conspiracy into this. Instead, it's more likely that Stupp and his team was familiar with temozolomide due to collaborations and discussions at Schering-Plough.

Schering-Plough was a US.-based pharmaceutical company that in 2009 merged with Merck.

Regardless of the reasons that temozolomide was selected, it has become part of the Stupp protocol and is widely-used for this reason. The Stupp protocol can be summarized as follows:

  1. Concomitant Radiotherapy/Chemotherapy. Starting about 5 weeks after a glioblastoma diagnosis, the patient receives one 2 Gy (Gray) dose of focal, megavoltage X-Ray radiation. "Focal" irradiation, where the beam is restricted to the resection bed, contrasts with "whole-brain" irradiation. In principle, focal irradiation is safer because those parts of the brain not affected by tumor are not irradiated. The patient receives a total of 30 of these doses Monday through Friday, holidays excepted. Every day (including weekends and holidays) during this phase, the patient takes a single oral dose of temozolomide. The dose is calibrated for the patient's surface area.
  2. Break. Four weeks during which no radiation or drug is administered.
  3. Adjuvant Chemotherapy. The patient receives a single oral daily dose of temozolomide for seven days, followed by 21 days of no temozolomide. The daily dose is roughly double what it was during the Concomitant phase. Up to six one-month cycles may be performed, depending on patient tolerance.

The dose of radiation given in the Concomitant phase is worth looking at more closely. The quantity "2 Gy" may not mean much unless you're an oncologist or radiologist. It helps to put this number into perspective. According to a Safety Card published by the American College of Radiology, 2 Gy is roughly equivalent to 1,250 brain CT scans (without contrast). Put another way, 2 Gy is equivalent to the dose you'd receive after 8,700 months of exposure to the Earth's background radiation. So glioblastoma patients following the full Stupp Protocol receive a non-trivial amount of brain irradiation.

Now for the "dismal" part. The Stupp protocol, despite its position as a cornerstone of modern glioblastoma treatment, is not very effective. It does appear to be better than nothing and alternatives that are no longer used, but that's a very low bar.

Stupp himself has always been up-front about the limitations of the protocol that bears his name. For example, the 2005 paper notes:

At a median follow-up of 28 months, 480 patients (84 percent) had died.

That's not a quote describing some long-abandoned approach. The quote describes today's standard of care, used in hospitals today around the world. It describes the very protocol I was about to subject myself to.

A more detailed picture of likely patient outcome comes from the metric "median survival time." Median survival time is the time required (since diagnosis) for half of the patients in the study to die. Stupp's team reported a median survival time for patients on the protocol of 14.6 months.

Once again, it's important to put this number into context. Given that all treatments have side effects, and that this applies especially to radiation and chemotherapy, the number that would be most helpful is median survival without any treatment. Unfortunately, Stupp's paper does not report the median survival of patients who did not undergo the Protocol.

However, we can get an idea of this number from previous work. A 2005 paper from a different group reported a median survival time of 4.9 months for a group of glioblastoma patients. So the Stupp Protocol (consisting of both Concurrent and Adjuvant phases) would appear to add about 10 months of survival time for a typical patient.

How much of this benefit can be attributed to radiation therapy by itself? That's a much harder question to answer. As the Stupp paper notes:

This trial was designed to determine whether the addition of temozolomide to radiotherapy early in the course of treatment prolongs survival among patients with glioblastoma, but it was not designed to distinguish between the effects of concomitant therapy with radiotherapy plus temozolomide and adjuvant treatment with temozolomide. At the time the trial was conceived, it was deemed most important to administer chemotherapy early in the course of the disease, for a sufficiently long time, and concurrently with radiotherapy. …

Even so, the picture is more complicated still. Recall that temozolomide's mechanism of action is DNA alkylation. Brain cells don't just take that kind of chemical abuse without fighting back. They can and do produce a wide range of enzymes whose purpose is is to undo the very damage caused by drugs like temozolomide. One of them, MGMT, is carried in abundance by the majority of glioblastomas. The enzyme can be silenced through "methylation" of its corresponding promoter region, but this only occurs in a fraction of patients.

This molecular nuance, MGMT promoter methylation, has very important practical consequences. In a companion paper, Stupp's team aimed to fill in this picture, concluding in the abstract:

Patients with glioblastoma containing a methylated MGMT promoter benefited from temozolomide, whereas those who did not have a methylated MGMT promoter did not have such a benefit.

However, the text of the companion paper walked this blanket statement back a little, noting:

… patients whose tumors are not methylated at the MGMT promoter appear to derive little or no benefit from the addition of temozolomide to radiotherapy. For these patients, alternative treatments with a different mechanism of action or methods of inhibiting MGMT should be developed.

As with all things in medicine, there are seldom black-and-white answers. The question of MGMT methylation is a case in point. It's often presented as an on/off kind of phenomenon, but gliomas are notoriously heterogeneous. Some cells within the same single tumor can show MGMT promoter methylation, whereas others do not. Instead of acting like it's connected to a switch, MGMT methylation behaves more like it's connected to a dimmer. This is one reason that some doctors recommend temozolomide during both concurrent and adjuvant phases of the Stupp protocol — regardless of MGMT promoter methylation status. As my own Dr. Staff Surgeon put it to me: "I prefer to let the tumor tell me whether or not it responds to temozolomide." In other words, until something better comes along, take the temozolomide — even if your tumor's promoter is "unmethylated." The flip side of this is that methylation does not necessarily imply better outcome.

The current standard of care, the Stupp Protocol, suffers from many limitations. It's crucial that glioblastoma patients and those who care about them understand and accept these limitations. Glioblastoma is not a cancer you beat; it's a cancer that will kill you unless something else does first. An important reason for this dismal outlook is that the standard of care is stuck in the 1980s.