ADCs Are Poised to Enter HER2-Mutant NSCLC Armamentarium

Antibody-drug conjugates (ADCs) are on the cusp of being introduced into the treatment paradigm for patients with HER2-mutant non–small cell lung cancer (NSCLC), said Paul A. Bunn, Jr, MD, who added that agents, such as fam-trastuzumab deruxtecan-nxki (Enhertu), have demonstrated unprecedented responses for this patient population where there are currently no approved targeted therapies.

“We now have a least 1 ADC that [can elicit] objective response rates (ORRs) higher than anything else reported for activating HER2 mutations [in NSCLC], with very long progression-free survival (PFS) rates similar to TKIs," said Bunn. “Albeit, these agents have drugs attached to them, so the [associated] adverse effects (AEs) are higher than they are for driver TKIs. They are, however, lower than with chemotherapy because the drug is attached to the antibody and internalized. We hope that agents like trastuzumab deruxtecan are approved in the near future for first-line use in patients with HER2-mutant NSCLC.”

During the 2020 ASCO Virtual Scientific Program, interim results from the phase 2 DESTINY-Lung01 trial revealed that trastuzumab deruxtecan induced a confirmed ORR of 61.9% (95% CI, 45.6%-76.4%) with an estimated median PFS of 14 months in patients with relapsed/refractory HER2-mutant NSCLC.

Additionally, the toxicity profile of the ADC was found to be consistent with previously reported data.

In an interview with OncLive® during the 21st Annual International Lung Cancer Congress®, Bunn, a distinguished professor of medicine, James Dudley Chair in Lung Cancer Research, Division of Medical Oncology, University of Colorado, and a 2014 OncLive Giant of Cancer Care® in Lung Cancer, discussed the mechanism of action of ADCs, how targeting HER2 mutations may impact the NSCLC paradigm, and exciting ADCs poised to enter the space. 

OncLive®: Could you provide an overview of the emerging role of ADCs in the HER2-mutant NSCLC space? What is the mechanism of action of this class of agents?

Bunn: ADCs are a new form of cancer treatment. Each of these molecules consists of 3 elements: a monoclonal antibody, a toxin, and a linker, which chemically links the antibody to a drug or toxin.

How effective any ADC is, depends on several variables. How is the antibody binding to the antigen? How is the antigen expressed in cancer cells? Is it expressed on every cancer cell or only some? Is it expressed on normal cells or only cancer cells? When an antibody binds to the antigen, what happens to the antibody? Is it internalized and brought inside the cells, or is the antigen recognized, shed, and not brought into the cell?

Obviously, what we would like is to have an antigen that is on every cancer cell, but not expressed on normal cells. We’d also like to have an antigen that is internalized to bring the drug inside the cell upon binding.

Then there is the linker. We want the linker to be very stable so that when the ADC is circulating in the blood, the toxic payload is not released into the circulation. The linker should be stable in the circulation, but easily cleaved when the ADC is internalized into the cell.

The third element, of course, is the payload. This payload could be any form of toxic drug. Obviously, many of the adverse effects (AEs) will be the same as [those associated with] the drug because not all of it will be internalized into the cell.

A very important property is how many drug molecules are attached to each antibody molecule. Some ADCs only have 1 or 2 molecules of the payload, but some have up to 10 drug molecules per antibody. Those may be more toxic, but they also may be more effective at [delivering] the drug inside the cancer cell.

Trastuzumab deruxtecan is an exciting agent in this space. What data have we seen regarding that agent, and how do the interim results of the DESTINY-Lung01 study stack up against current standard of care options?

There are no readily available ADCs that are approved and routinely used in any stage of lung cancer. However, there are a number that are being tested in advanced stage lung cancer. Trastuzumab deruxtecan is an ADC where the antibody is the anti-HER2 antibody trastuzumab (Herceptin). The drug is an irinotecan derivative.

The ADC has been studied in 2 groups of patients: those who have HER2-activating mutations, the majority of which express HER2 on the cell surface, and those who do not have HER2 mutations, but do have high levels of HER2 expression on the cell surface determined by immunohistochemical staining with an anti-HER2 antibody to be 2+ or 3+ in intensity. 

At the 2020 ASCO Virtual Scientific Program, the results of 42 patients with HER2-mutant NSCLC were presented. The ORR was 62% and median PFS was 14 months [with trastuzumab deruxtecan]. The majority of these patients had received prior treatment.

With a standard drug in this setting, you would expect to see a response rate of 10% and a PFS of 3 months. The [findings] from DESTINY-Lung01 were way better than we could have expected, and actually better than anything else in this setting. These are perhaps the best ADC results that have been presented to date in lung cancer.

These are very exciting findings and will undoubtedly lead to additional studies. Hopefully, they will also lead to approval of this agent in the future, which would likely be the first ADC approved in lung cancer.

If trastuzumab deruxtecan is approved, how will it affect the treatment paradigm for patients with HER2-mutant NSCLC?

At the present, there is no approved drug for HER2-mutant NSCLC. The standard treatment for those patients is platinum-doublet chemotherapy. Even in the first-line setting, platinum-doublet chemotherapy has a response rate of about 30% to 40% and a PFS of about 6 or 7 months.

Both the 60% ORR and 14-month PFS rate with trastuzumab deruxtecan are higher than any platinum-doublet chemotherapy. One would anticipate, because historical controls can be used for accelerated approval in some instances, that if this agent were on the market that its primary use would be in the first-line setting for patients with HER2-mutant NSCLC, which accounts for about 2% of all NSCLC.

What other ADCs are poised to enter this space?

There are several other ADCs being developed where Trop-2 is the target. Trop-2 is an antigen that is expressed more so on cancer cells than on fetal cells, but not much on normal tissue cells or lung epithelial cells. A very similar [irinotecan-derived] ADC with a similar linker was studied in advanced NSCLC, showing an ORR in the 30% range as opposed to the 60% range. As I mentioned before, in this setting, chemotherapy has an ORR of about 10%. Therefore, [a] 30% [ORR] is viewed as a major advance. [Responses] did not last 14 months as with trastuzumab deruxtecan, but lasted 5 to 6 months. In the advanced disease setting, that looks promising.

Also, there are some ADCs [against] AXL, which is highly expressed in many lung cancers. Again, responses were seen [with those agents].

These are the first 2 [ADCs] with which we’ve seen fairly exciting results. Although they are not ready for primetime at the moment, my hope is that within a year, trastuzumab deruxtecan may become commercially available and the others will enroll in ongoing phase 3 trials versus standard therapy.

The HER3-targeted ADC U3-1402 showed promising preliminary results. Could you speak to the mechanism of action of that agent?

One molecular target which was recently identified but for which there is no approved agent is NRG1fusions. NRG1 fusions activate HER3. Therefore, there is a bispecific monoclonal antibody [directed at] HER2 and HER2. In preclinical models, this bispecific antibody had very good effectiveness in NRG1-mutant patients. The group from Memorial Sloan Kettering Cancer Center reported objective responses in patients with NRG1 fusions using this antibody. Albeit, we are talking about single-digit numbers, but there have been responses. 

Several agents are being testing in NRG1 fusion–positive patients. There is tarloxotinib, [ADC U3-1402], and there are HER3 antibodies that are being tested in the subgroup of patients. It is very early, but the toxicity seems to be acceptable [with these agents]. Preclinically they work. 

Where should future research be focused in the space?

Most of the activating genetic alterations have been described. There may be some [alteration] that occurs in 0.2% of patients and could be identified, but I would say that we have identified the vast majority of driver alterations. Most of the TKIs for these driver alterations have ORRs in the 60% to 75% range and PFS rates ranging from 1 year to up to 3 years depending on the driver and the drug. However, these agents don’t cure patients. The question is, if you don’t kill [enough] cancer cells and other cancer cells persist, it’s inevitable that patients will [develop] resistance.

Then the question becomes: Why are these cells persisting in the presence of an inhibitor when they express the target? We don’t know the answer to that. It is some form of senescence. When the cells senesce, they go to sleep; they do not grow or die. Eventually, they will lead to resistance. To me, the biggest issue for driver alterations is determining the combination that is going to allow us to get complete remissions and cure some of these patients. That is an elusive goal, but one hope I have is that these neoadjuvant studies of targeted agents in early-stage disease will be able to tell us why the cells are persisting and what senescence mechanisms are allowing them to survive.

Reference

Smit EF, Nakagawa K, Nagasaka M, et al. Trastuzumab deruxtecan (T-DXd; DS-8201) in patients with HER2-mutated metastatic non-small cell lung cancer: interim results of DESTINY-Lung01. J Clin Oncol. 2020;38(suppl 15):9504. doi:10.1200/JCO.2020.38.15_suppl.9504