Round 3: What Is Needed to Establish DLL3 as a Clinically Actionable Biomarker?

In the final round, the discussion shifts from opposing positions to the path forward. Faculty explore the clinical evidence, assay and scoring standards, specimen requirements, reporting practices, and pathology–oncology workflows that would be needed for delta-like ligand 3 results to become reliably interpretable and potentially useful in treatment or clinical trial decisions. 

To hear arguments related to DLL3 predictive utility, see Round 1. 

To hear arguments related to who to test and when, see Round 2. 

To view the full series, click here.

Transcript

Jonathan Strosberg, MD: All right. So what is needed to establish DLL3 as a clinically actionable biomarker? In other words, what level of assay scoring, threshold standardization is needed? Should threshold be universal or therapy-specific? And what clinical evidence and workflow changes are needed to move DLL3 from an emerging biomarker towards routine predictive use? So Dr Roden is going to wrap things up here. 

Anja Roden, MD: So what does the pathology or the pathologist needs to produce a reliable clinically interpretable DLL3 results? So first of all, assay standardization. We talked about it. Currently, most of the trials and the publications use the SP347 from Ventana. And given our PD-L1 experience, it would be really good if we could just have 1 assay. Now, I realize we may not be able to have 1 threshold for all tumors, so there might be different thresholds. But if we, for instance, could just report the percentage of tumor cells that express DLL3 and the intensity, maybe an H-score, that gives you, as clinicians, all the information you need to put patients on different trials, even if they have different cutoffs. 

Even though we don't know if DLL3 is the predictive biomarker right now, in the lab, we have to treat it as predictive biomarker. It's not a diagnostic biomarker for us, it's a predictive biomarker. That means in the US we have to follow the CAP guidelines for validating this marker. You need 20 positive and 20 negative examples. And then we have to report it as a predictive biomarker. 

How do we report predictive biomarkers? The report has to include the clone and the company that provided this clone. We have to report our scoring, so it means percentage of tumor cells expressing DLL3, cytoplasmic and/or membranous. We have to report the preanalytic conditions that this was validated only on formalin-fixed paraffin-embedded tissue, not on decalcified tissue or any other fixatives. And that the specimen went into formalin within 1 hour of acquisition of the specimen and was fixed for 6 to 72 hours. And then again, for who needs the testing and so on, really requires the discussion with the oncologist. 

Nitya Raj, MD: Okay. And I'm going to speak to what oncology needs to use a DLL3 report for treatment or trial decisions. Okay. So I think that really the summary here is that as oncologists, we need results that are quantified, reliable, and they're relevant. We have to be able to take that result and say, okay, what can I offer my patient as a next choice here? 

So when we think about the results, and a lot of this we've already touched on, I think what's going to be helpful for us is understanding the percentage of tumor cells that are positive. A lot of the studies right now are using that as a defined cutoff, as well as, I think, staining intensity is certainly going to be a factor. And then what's the threshold? So that's what we're going to need to understand. And then in terms of the reliability, what's the success of the test, the adequacy of the specimen.   

I think one of the most challenging things for us, and I'm sure you all remember this from the early days of next-generation sequencing and still the struggles we have now with the tumor tissue, is just when you get those reports that are like, quantity not sufficient or quality not sufficient. It was incredibly frustrating when we were waiting a few weeks, which I think is longer than this test takes, and then you get kind of that result back. 

And then finally, applicability, the assay, the platform, the scoring method and how it aligns with that relevant therapy or whatever clinical trial you may be looking at, what the criteria are for that study. So when we think about the pathology report in some, it should make the DLL3 result interpretable, clinically useful. But the challenge here is that the prospective evidence that is hopefully coming needs to establish whether that DLL result is going to predict benefit for a specific DLL3 targeted therapy. 

Dr Strosberg: Okay. Thanks. Appreciate it. All right. So what must pathology and oncology resolve together before DLL3 results can reliably guide patient selection? I mean, I think we kind of covered it. I don't know if either of you have anything else to answer. I mean, we need more standardized pathology reports and I guess more clinical trials, but do you have anything to add? 

Dr Roden: No. And just whom do we have to test? That's... 

Dr Raj: Yeah, I think that's the big question. We were chatting about it even before this session. I think at some of the bigger academic centers now we're struggling through understanding, and we advocate for reflex testing in some cohort. We're not doing it standard either at Memorial. As Dr Roden said, they aren't doing it standard at Mayo yet. And we're trying to think through this to facilitate having that information as early as possible and utilizing the tissue well, kind of at diagnosis or presentation to the institution. 

Dr Strosberg: Yeah. Maybe I'll ask 1 more question. Let's say there was a phase 3 trial of DLL3 targeting drug for epNEC. Would you advocate at this point having a certain threshold? Or would you say it just has to be positive or take all comers? How would you design that? 

Dr Raj: So I think that it's a great question. I mean, based on some of this early data, just using the DLL3 agent, we're seeing this greater than 50% expression being potentially important. I think we have to weigh that against not having a lot of therapies in this disease and the off-chance that there are a handful of these lower-expression tumors that perhaps are getting benefits. So, as someone who's trying to offer as many options for patients when there are not that many options, I love the idea of including all epNECs. I can appreciate the argument for cutting off at the 50% expression. 

Dr Strosberg: Yep. 

Dr Raj: I don't know how you feel about it, but... 

Dr Strosberg: Well, for sure, I would exclude 0%. I think we can all agree on that, even though that's obviously not used in small cell lung cancer. But I mean, you would learn more if you took all comers and you could have co-primary endpoints for all comers and DLL3 above 50%. But at the same time, it's a rare disease, and you don't want to dilute the population too much. So yeah, I'm not sure how I would design that. So DLL3 is a established expression marker and therapeutic target, but still unclear whether it's predictive. Its value varies by disease setting and intended use, so it's very common in small cell lung cancer, more heterogeneous in epNEC. And we need more evidence both from clinical trials, as well as a standardized pathology metric for evaluating DLL3. And so, I think future years will tell us to what extent DLL3 is actually predictive for extrapulmonary neuroendocrine cancers. So thank you all for listening. Appreciate it. 

All right. So maybe we can have a little discussion among the panel. So based on the current evidence, what can DLL3 expression reliably tell us today, and what additional evidence would establish predictive utility? And I guess 1 question perhaps for you, Dr Roden, is do you think percentage expression will ultimately be more important or H-score or any predictions on that? 

Dr Roden: So I don't think we can really answer this question yet without the clinical trials. I mean, we can, I would say, pretty reliable tell you the percentage of tumor cell expression, at least if we would use a 50% cutoff. I don't know about other cutoffs. We would have to actually look into that. And you also know there will be some inter-observer variability if we go for intensity. So my 1+ might be my colleague's 2+, or my 3+ might be my colleague's 2+. 

And on top of it, there will be variation in intensity even within 1 tumor. So, if you see a resection specimen in 1 area, you will see 3+. In another area, you will see 2+. So we can report all of that. Whether we can report it very reliably, needs to be studied. But what that really means as a predictive marker, we don't know until we really have these all in clinical trials, I would say. 

Dr Strosberg: Great. And let me ask you 1 more question. To what extent does DLL3 expression vary, you think, inter-tumorally and intra-tumorally if you happen to have more than 1 site? 

Dr Roden: So this other good question. There is 1 small study that looked at small cell lung carcinomas at the resection specimen and then at the recurrence, and there was about an 88% concordance. However, this is only 1 study, and we have not looked into the distinction between primary tumor and metastasis and primary tumor and then recurrence in extrapulmonary neuroendocrine carcinomas. And also the study on small cell carcinomas has not been validated. 

Based on my experience, we do see different intensities within 1 tumor in resection specimens. Now, mostly we get biopsies, so you can't really talk about that, but there will be variation, I'm pretty sure. 

Dr Strosberg: Got it. And Dr Raj, am I correct in saying that they're doing DLL3 imaging studies at Memorial? I think there's a PET scan that's being developed. 

Dr Raj: Yeah. I mean, there's a lot of interesting early work going into DLL3 PET scans as a means to kind of look by imaging at the distribution of the receptor. And I think it speaks to the question you just asked about heterogeneity, like amongst what we may see by IHC testing in the primary, the metastases at recurrence and whatnot. And so, there's some early work that we're doing at Memorial looking at expression on PET scans. And then the next step is to kind of look at that in the context of response to therapies targeting DLL. 

Dr Strosberg: Great. And what clinical evidence would convince you that DLL3 expression is predictive? 

Dr Raj: Yeah. So this is an exciting time for extrapulmonary neuroendocrine carcinoma. I mean, we have not had many therapies. We've had platinum-based chemotherapy, and then we extrapolate our second-line regimens from site of origin-specific options for non-neuroendocrine carcinoma disease. And so, for extrapulmonary neuroendocrine, the fact that we may have a biomarker that we might be able to look at and target is exciting. 

So some of the evidence we have already that we've talked about is very promising in terms of some of the response rates and PFS data that we're seeing. But what we need now is some of the bigger phase 3 data, which is hopefully coming soon, kind of in the next few years. But that randomized data taken together with some of the phase 2 data, specifically looking at different thresholds is what's going to help us decide what is truly predictive in terms of DLL3 expression. 

Dr Strosberg: Great. Thank you.