Expert Shares Obstacles With Neoadjuvant Chemo in GEJ and Esophageal Cancers

David Ilson, MD

According to recent findings, pre-operative chemotherapy administered to patients with esophageal cancer may not have a particularly significant benefit for those who are undergoing resection, says David Ilson, MD, PhD. However, researchers globally debate whether chemotherapy alone or combined with radiation therapy is the optimal treatment for patients, specifically in tumors of the esophagus and junction versus the esophagus and stomach.

OncLive: Can you tell us about your recent studies into adjuvant chemotherapy plus surgery in esophageal and GE-junction cancers?

In an interview with OncLive, Ilson, a medical oncologist at Memorial Sloan Kettering Cancer Center, discusses pre-operative chemotherapy alone versus chemotherapy plus radiation therapy for the treatment of patients with esophageal and gastroesophageal (GE)-junction cancers, and how recent data points to the latter being the better choice.Ilson: I recently presented a review on recent literature from a pre-operative treatment in esophageal and GE-junction cancers. It's pretty clear that these cancers, including gastric cancers, benefit from a combination of adjuvant chemotherapy plus surgery. Whether that chemotherapy is given before surgery, which is done typically in Europe and the United States, or surgery first and then chemotherapy after the surgery as it's done in Japan, there's evidence that combining chemotherapy with surgery offers a benefit.

What's been debated is whether or not there's an additional role for adding radiation therapy. That topic came up in a recent lecture about the role of the optimal treatment of pre-operative therapy for esophageal and GE-junction cancers. There are two approaches: some favor chemotherapy alone, whereas in the United States we're more convinced that radiation needs to be combined with chemotherapy, particularly in tumors of the esophagus and the junction between the esophagus and the stomach.

What I did was go back and look at older studies and compare them to data from two very large recent trials looking at chemotherapy alone. What's clear from these studies is that giving chemotherapy alone as a pre-operative treatment does improve survival, but you also have to worry about good surgical outcomes with these diseases. In order to cure a patient, you have got to get a clear margin resection or R0 resection, and that's an important endpoint of these studies—not only getting a curative surgery, a clear surgical margin, but also reducing local recurrence of the cancers.

If you look at older studies of perioperative chemotherapy without radiation, there was a large negative trial to the approach, which showed no benefit for adding 5 months of chemotherapy compared to surgery alone. That study shows that about 40% of patients could not be operated on, and among the people who were operated on, the risk of the cancer coming back locally was about 30%. So it represented quite a huge failure rate of this approach.

In Europe, there were early studies of pre-operative chemotherapy in esophageal cancer that showed marginal survival benefits of about 5%. It also showed this 40% rate of inoperability with chemotherapy alone and local failure rates of 30%. Local control with chemotherapy alone is quite poor. Based on the success of early trials combining radiation and chemotherapy for esophageal GE-junction cancers where you've got potentially better rates of negative margin resection, lower rates of local recurrence, and improved survival, this approach is increasingly been adopted in the United States.

The pivotal trial came from the Netherlands, which compared surgery alone versus a very short course of chemotherapy, weekly carboplatin paclitaxel, and radiation; that study showed improved survival. It took the curative resection rate of about 60% and bumped it up to about 90%, and survival was improved; the treatment also resulted in a reduction of this local recurrence rate from about 25% to 30% down to 14%.

The point of my lecture was to look at some recent studies from the United Kingdom that continued to pursue this chemotherapy-alone approach. One was OEO5, which was a 900-patient study of esophageal and GE-junction adenocarcinomas, and guess what? After chemotherapy alone, the R0 resection rate was about 60% to 65%, so no improvement over 20 years. This is despite the fact that patients on this trial were screened with endoscopic ultrasound and PET scanned, so you could argue that patient selection was not the problem here. They were staged with modern staging techniques, and still with chemotherapy alone, you had very poor rates of curative resection.

A subsequent study that was presented by the British, ST03, was focused more on stomach cancer but also included GE-junction cancers and allowed a small amount of esophageal cancers. Two-thirds of the patients on ST03 had either esophageal or GE-junction cancers, and only about a third had stomach cancers. This approach also looked at chemotherapy alone, and this was a 1000-patient trial; if you look at the rate of curative resection for esophagus and GE-junction subtypes, it was 67%. It was terrible.

My point was that 20 years have gone by, and while chemotherapy is important to deliver as part of pre- or post-operative treatment, chemotherapy alone for this anatomic subtype results in high rates of non-curative resection and high rates of local recurrence. This is the argument that chemotherapy should be included. There's not general agreement about this, because some would say "oh, we need to do more studies," but my point is that we just had data released on 1600 patients with R0 resection rates of 60% to 65%, which to me is unacceptable.

The last issue that came up was a gastric cancer trial looking at the role of post-operative radiation. This was a study mostly in gastric cancers; only about 17% had GE-junction, so the majority of the patients on this trial had gastric cancer, a cancer in which there's evidence that chemotherapy by itself may be adequate and that radiation may not be needed. This was a trial of patients getting perioperative chemotherapy alone for gastric and GE-junction cancers versus perioperative chemotherapy with the addition of radiation post-operatively. This study showed no benefit for the addition of radiation to chemotherapy. The verdict might be that "they're equivalent," but to me if I had the choice of giving a patient more chemotherapy or putting them through 5 weeks of radiation with no difference, I wouldn't give them radiation.

Do you see the treatment paradigm for gastric cancer shifting?

There are old American data that suggested chemotherapy and radiation therapy after surgery improved outcomes, but the problem with those studies was that the quality of surgery was quite poor. Additionally, in this American trial, the “detour resection,” which is a standard of care for curative gastric cancer, showed a benefit of radiation only in 10% of patients. The vast majority had sub-optimal surgery, and the argument is that the benefit of radiation in this older study was to make up for inadequate surgery.I don't think it's going to shift, because the use of perioperative chemotherapy alone for gastric cancer alone is an accepted standard. I think it further questions the role of radiation in that subgroup. To me, I think there is increasing acceptance of adding radiation preoperatively to the tumors that are higher up, particularly given these data with 1600 patients with very poor surgical outcomes, and data that may improve surgical outcomes to enhance curative surgery.

Also, the approach of chemoradiation is a very short treatment with a very limited exposure to chemotherapy, and it provides an ideal opportunity to add new therapies. That's the other argument I make. We keep discussing that we should do more studies, and we're essentially doing studies on treatments that offer small benefits. To shift around these modalities to change the outcome by 1% or 2% doesn't really warrant larger studies. We need to accept certain treatment paradigms and build on them.

The future is going to be in more specific, targeted therapies. HER2-targeted agents are being studied with chemoradiation and chemotherapy in the perioperative setting, and that's an important advance. We're going to see the introduction of immunotherapies in controlled studies as part of adjuvant chemotherapy and chemoradiation. We’re also going to be looking at better markers of response to chemotherapy; PET scan imaging was recently evaluated in a US trial, which actually used the scan to either continue or change treatment pre-operatively depending on the response to the scan.

Molecular markers to response for chemotherapy haven't really been established, with the exception of HER2. A number of other targeted agents that we thought were going to work in more advanced disease like MET inhibitors or EGFR inhibitors really did not survive evaluation in phase III studies. HER2 is really the only validated biomarker for introducing a new drug in metastatic disease. Determining whether introducing HER2-targeted therapy improves locally advanced disease will await the results of ongoing studies.

It's an exciting time. We are generating a lot of genomic profiling data, and we have a much better understanding of the molecular biology of different subsets of upper GI cancers. There appear to be four molecular subsets, and what we don't know is how to exploit those subsets therapeutically. How can we design therapies that might take advantage of these different molecular profiles and biologies of these different subtypes? That's a work in progress.