Yelena Y. Janjigian, MD
There is a great need for novel treatments for patients with HER2-positive gastric/gastroesophageal junction (GEJ) adenocarcinomas, a patient population that makes up approximately 30% of those with the disease, says Yelena Y. Janjigian, MD, a medical oncologist at Memorial Sloan Kettering Cancer Center.
The search for novel treatments is even more important considering the recent negative results of the phase III LOGiC trial, which looked at the addition of lapatinib (Tykerb) to capecitabine (Xeloda) and oxaliplatin (Eloxatin) for patients with HER2-amplified gastroesophageal adenocarcinoma.
Median overall survival (OS) in the lapatinib combination arm was 12.2 months (95% CI, 10.6-14.2) versus 10.5 months (95% CI, 9.0-11.3) in the capecitabine and oxaliplatin–only control arm (HR, 0.91; 95% CI, 0.73-1.12). Median progression-free survival (PFS) in the lapatinib and control arms was 6.0 months (95% CI, 5.6-7.0) and 5.4 months (95% CI, 4.4-5.7), respectively (HR, 0.82; 95% CI, 0.68-1.00; P
In an interview with OncLive
, Janjigian discusses possible causes of the negative LOGiC trial results, HER2 as a driver in esophagogastric cancer, and what the future holds for lapatinib. She also discusses the potential of other treatments for HER2-positive gastric/GEJ cancer, including novel HER2-targeted agents, VEGF inhibitors, and immunotherapies.
OncLive: What did we learn from the phase III LOGiC trial? Is there an understanding as to why lapatinib failed to show a survival benefit?
: Based on the trial results, as they stand right now without any additional sequencing data or biomarker results, we clearly do not have a reason why lapatinib failed both in the first- and second-line setting. Based on my experience, I think we need to look at lapatinib as a drug, the trial design, and HER2 as a target in gastric cancer, in order to understand why it failed.
Lapatinib is an oral agent, and certain patients may not have the same levels of the drug built up in their system. Patients who have had gastrectomies, for example, where the drug quickly goes through their system and doesn’t get absorbed, may not benefit from oral drugs, such as lapatinib. Patients who are able to get a full dose of lapatinib without any problems with absorption may have a better benefit.
There are also differences in HER2 expression. HER2 positivity is defined by IHC 2+ and 3+. If IHC 2+, then you perform FISH. Therefore, for patients whose tumors have a high level of HER2 positivity, they may a have a bigger benefit. All of that data needs to be carefully reviewed. Patients who have a high level of HER2 positivity, perhaps IHC 3+ and FISH-positive, may benefit from HER2-directed agents, particularly beyond trastuzumab (Herceptin).
Finally, what it comes down to is that lapatinib may not be sufficient for HER2 inhibition in gastric cancer. We may just need a stronger HER2 inhibitor or HER2 inhibition with other targeted agents or chemotherapy. Gastric cancer is not only HER2 driven, even in HER2 disease.
HER2 amplification by FISH was selected as a biomarker for the LOGiC trial, regardless of HER2 IHC analysis. What questions remain regarding the use of HER2 as a driver for HER2-targeted therapies?
The question everyone asks is, “Would a second study in a population that had high level of HER2 amplification be worthwhile with lapatinib?” I think the answer is, “No.”
We know enough from looking at this disease that lapatinib is probably not going to be sufficient by itself. I wouldn’t subject another 300 patients to chemotherapy plus lapatinib, even in a highly HER2-amplified population. This is because, in my experience and now emerging data suggest, combination therapies are needed. We need to look at combinations of HER2-directed agents, HER2 agents with VEGF inhibitors, and HER2 agents with PD-1 inhibitors.
That is the wave of the future. We’ve got to go forward instead of trying to correct the mistakes we’ve made in trial design and not subject another patient population to a potentially small benefit or a negative trial.
Are any ongoing trials looking at VEGF inhibition in this patient population?
There was a small phase II trial looking at FOLFOX plus bevacizumab (Avastin) and trastuzumab in first-line. That trial, even though there were only data on 35 patients reported, was unprecedented compared to historic control in response rate, OS, and PFS in that patient population.
Bevacizumab is not FDA approved in gastric cancer, and it is not going to be developed further because there was a large negative AVAGAST study. However, ramucirumab (Cyramza) is a VEGFR2 inhibitor that is FDA approved in gastric cancer. We are opening a trial with Eli Lilly that I am the primary investigator for. It is a Memorial Sloan Kettering Cancer Center trial right now but, if the results look promising, it will become a multicenter trial. In this trial, we are studying ramucirumab with trastuzumab and chemotherapy in the first-line setting.
What about combinations with immunotherapies? Are there any upcoming trials investigating PD-1 inhibition in gastric/GEJ?
We have a trial that will be opening soon and is going through the approval process. It is exploring the possibility that perhaps we should combine PD-1 inhibitors with HER2-directed agents.
We are looking at pembrolizumab (Keytruda) plus trastuzumab and chemotherapy in HER2-positive patients in the frontline setting.
Is there a possibility of PD-1 being used as a single agent in gastric cancer?
There is a large research program to use pembrolizumab as a single agent in pretreated patients in the second- and third-line settings. Across the board, the response rate in unselected patients is probably going to be in a factor of 15% to 20%; that is what the data suggest with nivolumab (Opdivo), pembrolizumab, etc. This is not going to be good enough.
Of course, it is a good place to start if you have a deep and meaningful response in 15% of patients. The responses are going to be more dramatic in hypermutated gastric cancer, where patients have hundreds and hundreds of mutations in their tumor. HER2-positive tumors are not hypermutated, generally. My sense is that PD-1 inhibitors alone will generally not be sufficient in our patients, and we need to capitalize on the HER2 in the tumor and probably do combination therapy.
What potential role could other HER2-targeted agents, besides trastuzumab, have in this space?
In the second-line setting, T-DM1 (Kadcyla) has failed. T-DM1 was positive in patients with breast cancer and everyone was really excited to see the data, but the strategy failed. I think it is because, at the time of trastuzumab resistance in gastric cancer, loss of HER2 and loss of expression on the receptor is a really big problem. T-DM1 still capitalizes on the drug’s ability to bind to HER2.
Therefore, if you don’t have HER2 or if HER2 has been altered, you are unable to strongly bind to the receptor. Thus far, for patients who are still HER2-positive at the time of trastuzumab progression, nothing is currently available.
There might still be a role for T-DM1 in combination. The first step would be to look at it in combination with trastuzumab in the first-line setting.
We are also planning to conduct a large intergroup study investigating paclitaxel plus afatinib (Gilotrif), an irreversible EGFR, HER2, and HER4 inhibitor. EGFR signaling, even though cetuximab (Erbitux) failed in the disease, has potential in at least some population of HER2-positive patients.
I am very interested in studying this in patients with trastuzumab-refractory gastric carcinoma. I have a small, but growing, phase II study of afatinib plus paclitaxel, with a targeted accrual of 46 patients. We have seen some patients with quite dramatic responses. Patients have remained on study for 1 year, so they were long responses. This is not a common occurrence in gastric cancer.
Based on this preliminary data, we are interested in exploring this in the second-line setting, comparing paclitaxel and afatinib with paclitaxel and ramucirumab, which is the standard of care. The idea is to determine if capitalizing on HER2 in the second-line setting is still a viable option because, so far, experience with lapatinib and T-DM1 tells us that perhaps it is not.
My sense is that it is probably not the case. Most experts would agree that part of the reason these trials failed is because they were studying the wrong patient population. HER2 testing is often done on archival samples, which can be 4 years old; you can’t really say that is sufficient. You need to biopsy the patient to determine if they are actually HER2 positive.
Janjigian YY. Lapatinib in gastric cancer: what is the LOGiCal next step? [published online December 23, 2015]. J Clin Oncol. doi: 10.1200/JCO.2015.64.2892.