The novel photosensitizing substance 5-aminolevulinic acid (5-ALA), which causes gliomas to fluoresce, was more effective than conventional microsurgery with white light in aiding the detection and removal of locally invasive brain tumors in a multicenter Phase III study by the ALA Study Group. As reported by German researchers in a study in Lancet Oncology (2006;7:392–401), this led to improved progression-free survival rates.
The study's coordinating investigator, Walter Strummer, MD, Vice-Chairman of the Department of Neurosurgery at the University of Düsseldorf, called 5-ALA “the first technical adjunct to be put through a truly randomized study.”
Better Tools Needed
Less than 20% of patients achieve complete resection of contrast-enhancing tumors with frameless stereotaxy or intraoperative magnetic resonance imaging, he and his colleagues noted, suggesting that surgeons using these tools have trouble discerning where the tumor stops and healthy tissue begins.
“In malignant glioma surgery, surgeons need technical adjuncts,” Dr. Stummer said in a telephone interview. “5-ALA is an adjunct that improves resections and benefits patients in terms of prolonged, progression-free survival. Also, in contrast to [intraoperative] MRI, 5-ALA is cheaper, can be used in a truly online fashion and is safe.”
5-ALA is a natural biochemical precursor of hemoglobin, he explained. The drug itself is not fluorescent, but causes fluorescent porphyrins to accumulate in cancerous tissue. A modified neurosurgical microscope is used to visualize fluorescence during surgery.
In the study, which was sponsored by medac GmbH, the manufacturer of 5-ALA, Dr. Summer, who is a paid consultant to the company, and his co-researchers randomly assigned 270 evaluable patients with suspected malignant glioma to receive either oral 5-ALA for fluorescence-guided resection (139 patients) or conventional microsurgery with tumor-enhancing white light (131 patients).
One to five grams of 5-ALA were dissolved in 50 mL of drinking water and administered three hours prior to the induction of anesthesia. All patients received fractionated radiotherapy 14 days after surgery.
After a median follow-up of 35.4 months, 65% of patients who received 5-ALA had their tumors completely resected, compared with 36% receiving microsurgery with white light.
Furthermore, the six-month progression-free survival rates associated with fluorescence-assisted surgery were twice that of white-light surgery (41% vs 21%, respectively). The median volume of residual tumors was smaller in the 5-ALA arm than in the white-light arm (0.0 cm3—range of 0.0 to 25.7 cm3—vs 0.7 cm3—range of 0.0 to 32.6 cm3).
The frequency of severe side effects reported within seven days after surgery was not significantly different between the two arms of the study.
The most severe adverse events during treatment (5-ALA vs white-light group, respectively) were hemiparesis (4 vs 2 patients), convulsions (3 vs 1) and epidural hematoma (1 patient in each group). Five patients in the 5-ALA arm and three in the white-light arm died within 30 days after treatment.
The study design placed no restrictions on therapy following disease progression, making an exact overall survival analysis impossible, noted Dr. Stummer. There was no significant difference in the estimated overall survival rates between groups, even when these groups were stratified by younger versus older patients.
Editorial: Cautious about Results
In an accompanying editorial, however, Frederick G. Barker, MD, Associate Professor of Neurosurgery at Harvard Medical School, was circumspect regarding the study results.
“This is evidence that better resection of glioblastoma probably does bring a modest prolongation of survival if done without increasing neurological deficits,” he said in an interview.
In the editorial, Dr. Barker clarified that complete surgical resection of gliomas was strikingly dependent on favorable prognostic factors (i.e., patient age, performance status, and the distance of the tumor from functional brain tissue) rather than on adjunctive technology alone.
“There was an interesting influence of age and performance status on the degree of resection, and an influence of proximity to eloquent brain, something which has never before been demonstrated but was expected,” Dr. Stummer said.
“Both study arms were consequently balanced regarding these factors, so these factors could not have been the rationale for the resection gains derived by utilizing 5-ALA or any of the advantages gained for the patient in terms of outcome.
“When we left the originally defined study groups and re-stratified by resection status—no residual versus residual disease—and looked at survival, there was a large difference—five months—between both groups,” he continued. “We then performed multivariate analysis with age, performance status, ‘eloquent location’ and resection status as covariates, and we found an independent impact of age and Karnofsky performance score, as expected. However, the most prominent independent factor was resection status.”
In his editorial, Dr. Barker wrote that past studies have shown that successful resection surgery results in modest progression-free survival benefits for patients with glioblastoma multiforme, with complete resection associated with a median overall survival of 15 to 18 months, compared with 10 to 12 months after a subtotal resection.
Dr. Stummer cited a European Organization for Research and Treatment of Cancer trial that demonstrated that concomitant radiochemotherapy with temozolomide was more effective in patients who had undergone a complete resection. As such, it would stand to reason, therefore, that improving tumor excision would slightly increase survival rates, although this hypothesis remains controversial.
Potential Investigator Bias
Owing to the nature of the investigation, patients undergoing white-light surgery did not receive a placebo, Dr. Barker noted. “This left open the possibility of more aggressive surgery in the fluorescence-guided group because of its unblinded design, rather than through the effects of the drug's fluorescence.”
Patrick Y. Wen, MD, Clinical Director of the Center for Neuro-Oncology at Dana-Farber/Brigham and Women's Cancer Center, said he agreed with Dr. Barker's assessment.
“One has to be cautious in interpreting the results, since they are so much better than would have predicted from previous data, raising the issue of unintentional investigator bias since the surgeons were aware which arm the patients were on.”
Dr. Stummer responded that although surgical bias cannot be ruled out—“it is essentially a problem with all open-label studies and one that all surgical studies are faced with, and the ALA study is no exception”—he nonetheless found it highly unlikely that any surgeon would attempt to remove more of the tumor in the experimental arm rather than the control arm.
He noted that residual tumor volumes following surgery in the white-light group were similar to those reported by a single, experienced center in a 2001 study involving 416 patients. “We used this study to indicate that patients in our control group were not getting worse surgery than in other centers—and thus any systematic influence by study surgeons appears very unlikely,” he said.
Comparative Surgical Techniques
Dr. Barker speculated that if current “aggressive” techniques such as neuronavigation, which uses intraoperative stereotactic guidance to differentiate cancerous from healthy tissue, or intraoperative MRI had been used on the study's control arm instead of conventional microsurgery with white light, the results of fluorescence-guided surgery may have appeared less impressive.
Intraoperative MRI and neuronavigation already provide surgeons with the means for achieving a greater extent of resection, Dr. Wen said. “It is unclear whether the use of 5-ALA is superior to these other methods.”
Dr. Stummer said that neuronavigation is the only technique that has been assessed prospectively and that it has shown no significant benefit in terms of resection. Furthermore, neuronavigation was associated with brain shift—“the intraoperative distortion of anatomy compared with that seen on preoperative imaging.”
As for intraoperative MRI and intraoperative ultrasonography, two surgical adjuncts currently used to aid surgeons in the removal of malignant gliomas, Dr. Stummer argued that the former was “cumbersome and expensive” and that the latter was frequently associated with interruptions during surgery.
“It is hard to say whether MRI would have mitigated any results from the ALA study,” Dr. Stummer said. “MRI has not been put to a truly randomized test. Moreover, we know that 5-ALA-derived fluorescence is more sensitive than contrast-enhancement on MRI for demonstrating residual tumor.”
Overall Survival Not Stressed
Long-term overall survival rates could not be statistically assessed owing to the study's design, he said.
“After disease progression there were no restrictions to therapy and it turned out that patients were getting more repeat surgery in the control arm—this was a positive aspect of the study. Thus, effects on survival were mitigated and failed to reach statistical significance. In future studies this effect will have to be controlled for, to truly be able to say anything about the effects of surgery on survival.”
He and his coauthors still provided an estimate of overall survival, which translated into a modest overall survival benefit of 1.7 months for patients in the 5-ALA group.
Older patients (over age 55) in the 5-ALA group and white-light group had an overall survival time of 14.1 months (range of 11.7 to 16.7 months) and 11.5 months (8.8 to 13.7), respectively. In contrast, younger patients (55 and younger) in the 5-ALA and white-light groups had an overall survival of 18 months (range of 13.0 to 20.8 months) and 17.5 months (14.3 to 21.2), respectively.
“This was different from what I would have predicted,” Dr. Barker said. “Some have suggested, based on little hard evidence, that more extensive resections are more beneficial for younger patients than older ones, which is the opposite of what was found. But the trial wasn't designed to test this, and probably too much shouldn't be made of the result.”
Dr. Wen speculated that perhaps younger patients tend to have more aggressive surgery in general and so the addition of 5-ALA did not improve the outcome as much.
Age and Reinterventions
An exploratory analysis showed that older patients (n=168) underwent fewer re-interventions (e.g., chemotherapy and repeat surgery) following initial surgery than younger patients (n=83)—55% vs 82%, respectively. Older patients in the 5-ALA arm had a longer time to reintervention compared with patients in the white-light arm—10.2 vs 7.1 months, respectively.
In contrast, the time to reintervention for younger patients in the 5-ALA arm compared with patients in the white-light arm was not considered statistically significant—8.0 vs 6.7 months.
“Surgeons may have been more restrictive in older patients because they believe potential neurological deterioration to be less well tolerated in older than in younger patients,” Dr. Stummer said. “It makes one wonder why older patients are attributed worse survival chances in the first place. It appears that their entire therapy was biased—a self-fulfilling prophecy?”
Dr. Stummer said he found it interesting that the experimental group as a whole had fewer surgical reinterventions than the control group—30% vs 37%, respectively.
“In my opinion, the impact of better resections and prolonged progression-free survival made interventions at an early stage unnecessary,” he said.
A Small Step Forward
Despite some questions regarding the 5-ALA study, Dr. Wen was relatively optimistic about the results, stating that the study provides evidence that greater extent of resection may lead to better outcome and improved progression-free survival.
Dr. Barker said, “It's nice that better resections seem to extend progression-free survival, but disappointing that the benefits—both progression-free survival and overall survival—were rather modest. But this disease has long been known to be surgically incurable. I'd like to see a similar technique that works for low-grade gliomas, where one might hope for a more durable benefit from complete resections.”
Both Dr. Barker and Dr. Wen cited the need for follow-up data to confirm the research before 5-ALA is introduced into standard practice. As of yet, 5-ALA has not been approved in the United States or European Union for the purpose of enhancing malignant gliomas during surgery.
Dr. Stummer is currently planning a follow-up study in which newly diagnosed glioblastoma multiforme patients receive “best possible surgery” in addition to a new regimen of concomitant therapy using temozolomide and radiotherapy followed by adjuvant temozolomide. “We want to see if a combination of these factors is additive,” he said.