CME
Physicians: Maximum of 1.00 AMA PRA Category 1 Credit™
Released: July 25, 2023
Expiration: July 24, 2024
TROP-2 as a Therapeutic Target
Stephen Liu, MD:
Now we will focus on the ADC target antigens that are of interest in lung cancer. Dr Borghaei, could you please explain what TROP-2 is?
Hossein Borghaei, MS, DO:
TROP-2 (trophoblast cell surface antigen 2, also known as tumor associated calcium signal transducer 2) is a cell-surface protein that is thought to promote malignant potential in several different tumor types, including NSCLC. TROP-2 is a stem cell marker that is associated with cell transformation, regeneration, and proliferation through interaction with various ligands.6 It is an important target for cancer therapy because it is involved in various signal transduction pathways that are associated with cancer cell growth and tumorigenesis.7 For example, TROP-2 can bind IGF1, thus attenuating IGF1R signaling and downstream target genes.8
TROP-2 Overexpression in NSCLC and Other Cancers
Stephen Liu, MD:
It makes sense to use ADCs in cancers that express the target antigen. Dr Heist, is TROP-2 overexpressed only on certain types of cancers?
Rebecca S. Heist, MD, MPH:
TROP-2 is upregulated in a very broad range of epithelial cancers, although it also can have some limited expression in normal tissues (eg, breast, lung, stomach, colon, prostate, pancreas, and uterus), which may contribute to some of the toxicities seen with TROP-2‒directed ADCs. TROP-2 is overexpressed in the majority of NSCLC, and an early study on 87 patients with NSCLC showed significantly higher TROP-2 expression in squamous cell carcinoma (n = 37) compared with adenocarcinoma (n = 50; P = .018), although there is overexpression in both histologies.9 In another lung cancer study, TROP-2 overexpression was observed in 75% (150/201) of squamous cell carcinoma and 64% (172/270) of adenocarcinoma but only 18% (21/115) of high-grade neuroendocrine tumors.10
Stephen Liu, MD:
Dr Borghaei, when considering TROP-2 as a target for lung cancer therapy, which cancer type (eg, adenocarcinoma, squamous, small cell) is the most relevant?
Hossein Borghaei, MS, DO:
TROP-2 seems to have a differential role based on lung cancer subtype. High TROP-2 expression was associated with higher lung cancer‒specific mortality in adenocarcinoma but not in squamous cell carcinoma.10 In a study of 87 patients with NSCLC, high TROP-2 expression was associated with significantly shorter survival in patients with adenocarcinoma (11.5 months vs 17.3 months; P = .002) but not in those with squamous cell carcinoma (14.6 months vs 17.2 months; P = .276).9 A retrospective study of 331 patients with NSCLC associated TROP-2 overexpression with shorter overall survival (OS) for patients with lung adenocarcinoma (P =.028) but not for patients with squamous cell carcinoma. There were no significant progression-free survival (PFS) differences with TROP-2 overexpression in adenocarcinoma or squamous cell carcinoma.11 Thus, from a targeting perspective, adenocarcinoma may be more appropriate. Also, given the lack of available treatment options for patients with small-cell lung cancer (SCLC), I think it may be important to further explore the activity of TROP-2‒directed ADCs in this group.
Sacituzumab Govitecan-hziy: TROP-2‒Targeted ADC
Stephen Liu, MD:
ADCs targeting TROP-2 are an emerging treatment option for patients with advanced NSCLC, but none has been approved yet in this setting.
Sacituzumab govitecan-hziy is a TROP-2‒directed ADC composed of a humanized IgG1 antibody connected to SN-38, the active metabolite of irinotecan (topoisomerase I inhibitor), via a pH-sensitive linker. With a high DAR of 7.6, a unique conjugation method stabilizes the SN-38 payload and prevents the glucuronidation that contributes to the intolerability of treatment with irinotecan.12
The pH-sensitive linker allows release of the payload in the acidic tumor microenvironment, which is expected to contribute to a bystander effect in nearby cancer cells. Sacituzumab govitecan-hziy is approved for treatment of adult patients with metastatic triple-negative breast cancer who have received ≥2 prior systemic therapies, including ≥1 for metastatic disease, and for treatment of adult patients with unresectable locally advanced or metastatic hormone receptor‒positive/HER2-negative breast cancer who have received endocrine-based therapy and ≥2 additional systemic therapies in the metastatic setting. In addition, it is approved for the treatment of adult patients with locally advanced or metastatic urothelial cancer who have previously received a platinum-containing chemotherapy and a PD-1/PD-L1 inhibitor.13
Various clinical trials are underway for sacituzumab govitecan-hziy in patients with NSCLC, which will be discussed later in greater detail.
Datopotamab Deruxtecan: TROP-2‒Targeted ADC
Stephen Liu, MD:
Another TROP-2‒directed ADC is datopotamab deruxtecan, which is a humanized IgG1 monoclonal antibody connected to DXd (topoisomerase I inhibitor) via an enzymatically cleavable tetrapeptide linker.14 With an optimized DAR of ~4, preclinical testing of datopotamab deruxtecan demonstrated efficient delivery of the payload into tumors with acceptable safety. Multiple clinical trials are underway in patients with NSCLC, which will be discussed later in greater detail.
ADC Efficacy: The Importance of the Target Antigen
Stephen Liu, MD:
ADCs are targeted at least to some degree, and the part that provides the specificity for drug administration is the antibody.
Hossein Borghaei, MS, DO:
Yes, the whole concept begins with the targeted delivery of a cytotoxic payload to specific antigen expressing tumor cells. This requires not only binding of the antibody to the target antigen, but also the further step of internalization of the ADC into the tumor cell. From a purely scientific point of view, this is a targeted delivery of a chemotherapeutic payload.
Now, in the case of a target antigen that is widely expressed on the surface of the cancer cells (eg, TROP-2 in lung cancer), this may seem like broad activity, but that is not the fault of the drug design. Yes, there is a bystander effect, and the second generation of ADCs that we are dealing with now do seem to work a bit indiscriminately, but they are not necessarily designed to be like that.
Melissa L. Johnson, MD:
It could be argued that in the case of TROP-2‒directed ADCs, it is just chemotherapy and not so targeted after all.
Hossein Borghaei, MS, DO:
I understand what Dr Johnson says, and I agree that to some extent it looks like we are just giving chemotherapy, but I really do think the ADC concept is more about a targeted delivery of extremely toxic payloads that we otherwise would not be able to give to patients.
Stephen Liu, MD:
I think the truth probably lies somewhere in between. Dr Borghaei’s point is that on paper, this is a really elegant, rationally designed way to deliver chemotherapy, using a tumor-expressed antigen as a Trojan horse to introduce an extremely cytotoxic payload, killing only the cancer cells and leaving all the normal cells behind. However, to Dr Johnson’s point, the biomarker-to-target specificity has not turned out as expected. In breast cancer, trastuzumab deruxtecan was approved based on HER2 overexpression, but in lung cancer, patients with activating HER2 mutations saw the greatest clinical benefit. This may be due to enhanced internalization of ADC complexes, as seen in HER2-mutated lung cancer cell lines and patient-derived xenograft models.15 In some cases, it seems like the target antigen expression level does not matter, when theoretically it absolutely should matter. For example, patritumab deruxtecan showed activity not only in breast cancer cell lines overexpressing the target antigen HER3, but also in those harboring clinically observed HER3 mutations.16
Rebecca S. Heist, MD, MPH:
It is logical to expect that TROP-2‒directed ADCs would work better in a patient with higher levels of TROP-2 expression, but we are not seeing that correlation. Perhaps there is a certain minimal baseline level of expression, and with that you have enough for your drug to get to where it needs to go. This is not quite the same as c-MET‒targeted ADCs, where you actually do see different activity at differential levels of expression.
Stephen Liu, MD:
I agree it is logical to assume that ADC efficacy should directly correlate with target antigen expression, but that just is not what we see. There are no universal truths when it comes to ADCs, and they are not all as targeted as we would like them to be on paper. It can be a bit frustrating.
ADC Efficacy: The Importance of the Linker and Payload
Stephen Liu, MD:
All of the components that make up the ADC seem very important to the final functionality. Although the antibody and payload are the 2 most obvious, we also have to keep in mind the importance of the linker—should we not, Dr Heist?
Rebecca S. Heist, MD, MPH:
The linker is important because it keeps the whole complex from breaking apart in circulation. At the same time, the linker must be able to release the payload when appropriate once the ADC has been internalized into the cancer cells. The linkers in use are all very different, and they are proprietary, which adds a big wild card to how exactly these ADCs actually function in the body. We know linker biochemistry plays a very important role in the activity of ADCs, but there are great differences between the various linker types. There are cleavable linkers, noncleavable linkers, and pH-sensitive linkers that are all proprietary, each with a different set of properties that affect how the drug is internalized and released, either in the bloodstream or the target cell.
Stephen Liu, MD:
Dr Levy, when we think of first-line immune checkpoint inhibitors (ICIs) (eg, atezolizumab, cemiplimab, nivolumab, and pembrolizumab), although there are some differences in the studies and drugs, we consider them to be pretty interchangeable. Am I correct in assuming that that is not the case for the different TROP-2‒directed ADCs, even though they share the same target antigen?
Benjamin Levy, MD:
Yes, there are clear differences related to unique linkers, and the different payloads we are looking at now (govitecan and deruxtecan for the TROP-2‒directed ADCs) may have different sets of toxicities.
Clearly, the off-target effects may be different, and this is playing out now in the clinic (as we will discuss later), where we are seeing certain ADCs with more potential toxicities related to interstitial lung disease or stomatitis.
Melissa L. Johnson, MD:
I would like to add how fascinating it is that we can treat patients with different disease courses using various ADCs that employ the same payload. Today while in clinic, I saw a patient with HER2-mutated lung cancer who has received trastuzumab deruxtecan. Then I saw a patient with EGFR-positive NSCLC, second line, post osimertinib, who is receiving patritumab deruxtecan (HER3-directed ADC). Finally, I saw a patient with SCLC receiving ifinatamab deruxtecan (B7-H3‒directed ADC). These are all the same payload being targeted to 3 different antigens for 3 different lung cancer types.
The other thing to keep in mind is that the DAR varies between trastuzumab deruxtecan (~8) and the different TROP-2‒directed ADCs (~4 for datopotamab deruxtecan, 7.6 for sacituzumab govitecan-hziy), and clinically all 3 manifest with different adverse events (AEs).
Stephen Liu, MD:
That is a great point, as DAR influences not only efficacy, but also potential toxicity. When we consider predicting toxicity, it really is a function of the payload potency, the linker type, and the DAR. Even ADCs with many similarities can have significant yet subtle differences.
TROP-2 as a Predictive Biomarker
Stephen Liu, MD:
It is logical to think that the natural biomarker for an ADC would be the target antigen, but as discussed above, that is not always the case. Dr Borghaei, what do we know about TROP-2 as a predictive biomarker in this space?
Hossein Borghaei, MS, DO:
This seems to be one of those antigens whose activity does not depend on significant expression. I am reluctant to call something a TROP-2‒directed ADC when it does not matter what the TROP-2 level is. I think each ADC needs to be considered separately. I would use the example of CEACAM5, where investigators are trying to determine if there is a cutoff CEACAM5 expression level. A phase II study is evaluating tusamitamab ravtansine in patients with nonsquamous NSCLC and negative or moderate CEACAM5 expression and high circulating CEA.17 We simply do not know at this point if even 1% is better and if efficacy is independent of expression level. For TROP-2‒directed ADCs, it looks to me like we may have given up on a biomarker for patient selection because there is so much expression in all lung cancer cell lines.
Melissa L. Johnson, MD:
Just to play devil’s advocate, what if the CEACAM5 strategy is wrong? It is extremely difficult if you have to screen 20 patients to find 1 patient whose disease expresses CEACAM5 at a high enough level to enroll in the phase III CARMEN-LC03 study (NCT04154956; a comparison of tusamitamab ravtansine vs docetaxel in patients with metastatic nonsquamous NSCLC and high-expressing CEACAM5). What if such a high level of expression is not necessary for efficacy? It is much easier to enroll with an all comer strategy.
Hossein Borghaei, MS, DO:
But how many patients with NSCLC do you have to screen to find 1 ALK rearranged patient?
Benjamin Levy, MD:
Only approximately 5% of patients with NSCLC show ALK rearrangements, but available ALK tyrosine kinase inhibitors show a 60% to 70% response rate in these patients, so it is worth it. Retrospective analyses of ALK inhibitors have shown median OS of 4.3-6.8 years.18 They really have been a game changer by showing the impact of targeted therapy for patients.
Rebecca S. Heist, MD, MPH:
If there was a biomarker available for the TROP-2‒directed ADCs—and TROP-2 immunohistochemistry is not it—where you could get the response rate up to 60%, you would want that, right? If we have an all comers trial and get a 20% response rate, that is great, but I would prefer to be able to screen and get a response rate of 60% to 70%.
Melissa L. Johnson, MD:
Absolutely—but from a study design perspective, if you do not have a known biomarker, should you not design an all comers trial to determine efficacy? Otherwise you limit enrollment based on assumptions.
Benjamin Levy, MD:
Is it so bad that we do not have a biomarker right now for TROP-2‒directed ADCs? These drugs work, and I think in the future we will absolutely try to screen patients by an enrichment strategy to further improve efficacy. I think the bar is high. You are going to have to get a 60% to 70% response rate—or at least above 50%. Although we would like to use trastuzumab deruxtecan for lung cancer, I think we only have a nascent understanding of why it works in HER2 mutated lung cancer and not HER2-overexpressed disease. It has been described that HER2 mutations enhance the internalization of the ADC.15 We did not see a similar phenomenon in the preclinical data when giving a TROP-2‒directed ADC. I think we are just scratching the surface of our understanding of ADC function. I believe that we will get to a point where we can enrich patients, but for now I do not see such stringent patient selection panning out.
Hossein Borghaei, MS, DO:
I do not think there is a good enough excuse to say all comers should get the drugs. These are potentially toxic drugs, and the activity is not very convincing.
Melissa L. Johnson, MD:
Currently, for all the patients with driver negative NSCLC, we have a nonselective strategy (chemotherapy plus immunotherapy) in the frontline setting. We know that not all patients derive the same benefit from TROP-2‒directed ADCs. You cannot look at someone and know whether they will do well or not. There is a real need to understand the next line of therapy. Otherwise, we are repeatedly separating our patients into those who have sensitive or resistant cancers without really learning more about the heterogeneity.
Stephen Liu, MD:
Although there is some disagreement, I think you are all providing a similar message. What we have right now is a strategy that does not use selection. We certainly want to work toward a way to enrich our patient population in the studies to better learn how to personalize therapy.
Back
