Look, before I begin this post let me say that I’m really not trying to be a negative person when it comes to Avastin (bevacizumab; see here and here). It is a great drug in the appropriate setting, and is has been proven to prolong survival in first-line treatment of advanced NSCLC patients when combined with chemotherapy. Given this success, as well as proven success in treating renal cell carcinoma, breast cancer, and colorectal cancer, it is understandable that people would keep testing it in new situations. It’s just that all of the recent publications that try and expand the role of Avastin to new areas seem to have disappointing results. Which brings me to today’s topic.

A new online publication in the Journal of Clinical Oncology by Dr. David Spigel and colleagues from the Sarah Cannon Research Institute in Nashville, TN, describes the experiences from 2 phase II trials in small cell (SCLC) and non-small cell lung cancer (NSCLC) using Avastin in combination with chemotherapy and chest radiation. There had been reports of safety concerns in the past with this combination, specifically with regards to something called tracheoesophageal fistula, but this was my own first look at the data.

A tracheoesophageal fistula is a pathologic connection between the esophagus (food pipe) and trachea (windpipe). It doesn’t take a doctor to imagine why it might be a bad thing to empty everything you eat or drink directly into your lungs, but in case you weren’t sure let me assure you this is what we in the business call “a bad thing”. It is always serious and often fatal, although if caught early can be treated with a stent that covers the opening and prevents passage of material between the two sites.

In 2006-07 there were two phase II trials enrolling patients to investigate the role of adding Avastin to chemotherapy and radiation in limited stage SCLC and in locally advanced (stage III) NSCLC. Concurrent chemoradiation represents the standard treatment for both of these conditions, so the only major change was in adding the Avastin. The SCLC study enrolled 29 patients, but stopped early after an unacceptable level of toxicity. In contrast, the NSCLC study only enrolled 5 patients before having to stop. I think these studies represent a cautionary tale, but are also a positive testament to the safety measures put in place to stop trials if they end up causing too much toxicity.

The SCLC study tested Avastin in combination with carboplatin and irinotecan for 2 cycles, followed by the same combination with concurrent radiation to the chest to a dose of 61 Gy. After completion, the patients continued Avastin alone until disease progression. Of the 29 patients enrolled on the trial, 2 died from documented TE fistulas, which is much more than one would expect from chemoradiation alone. In addition, a third patient died from fatal bleeding into the lungs but no autopsy was done so it is unclear if this was from TE fistula. The trial was then halted to further accrual and this combination has not been investigated further. However, the authors also reported another patient with SCLC who received maintenance Avastin after chemoradiation who also died from TE fistula.

The NSCLC trial enrolled stage III patients to a regimen of Avastin added to carboplatin, pemetrexed, and concurrent radiation. After the CRT, the patients had two cycles of the same chemotherapy plus Avastin, and then were continued on the Avastin alone until disease progression. Of the first 5 patients accrued, 2 developed TE fistulas, an unacceptable high rate of an otherwise rare complication.

The high incidence of TE fistula with Avastin and radiation has been known for some time from these trials, although I did not know the details. There were some interesting elements from the specific cases that may lead us to determine why this occurred. For example, while you might think it was the Avastin given during the radiation that was at fault, none of these patients developed the TE fistula during their radiation treatment (except for the one patient who died of bleeding for unclear reasons). Instead the other 4 patients developed TE fistula after finishing chemotherapy and radiation and while receiving Avastin alone in the maintenance setting.

In addition, all of the patients had severe esophagitis (inflammation of the esophagus), which is common after chemoradiation but seems to have been quite severe in these patients. One of the 3 SCLC patients was hospitalized for esophagitis, and both of the NSCLC patients had such severe esophagitis that they developed strictures (narrowing of the esophagus from scarring) that required repeated dilations to open them back up. It appears that severe esophagitis, and quite possibly mechanical stretching of the damaged esophagus, may have predisposed the patients to TE fistula. Overall 2 of the 5 patients died from the complications and the remaining 3 are still alive with stents.

It has been long speculated that Avastin, which inhibits vascular endothelial growth factor (VEGF), a protein important to cancer blood vessel growth but also to normal vessel growth, might impair wound healing. This is indirect evidence that this may be true, perhaps inhibiting the esophagus from healing properly. Avastin also strengthens the side-effects of chemotherapy, and may have simply increased the damage done to the esophagus from the chemoradiation. It is not possible to know for sure, but for now it appears that Avastin should not be used in combination with chemoradiation to the chest and perhaps not with chest radiation at all. Indeed the FDA placed this on the label for Avastin in what is called a “black box” warning.

After this one I promise to stay away from Avastin for awhile, unless of course something good comes out!