Thank you so much for the introduction. I am really excited to kick off this seminar by exploring some different ways that we can optimize timely molecular testing for patients with early stage non-small cell lung cancer, and importantly, how we can better understand and interpret the results from these biomarker testing reports.

Diagnostic and Therapeutic Subtyping in NSCLC

Before we begin, I wanted to just take a very high-level view of the remarkable evolution of diagnostic classification of non-small cell lung cancer and how these innovations have really opened up new avenues for treatment.

Here we begin with the histologic classification of lung cancer into the different subtypes like lung adenocarcinoma, lung squamous cell carcinoma and others. This really remains the foundation of how we think about this disease.

In the past decade, we now understand that lung cancer is really driven by different oncogenic molecular alterations such as EGFR, KRAS and ALK fusions. We have developed highly effective targeted therapies against these drivers as well.

Of course, we are understanding the different immune phenotypes in lung cancer, which are primarily classified by PD-L1 expression to help guide different types of immune checkpoint therapies, which have been quite transformational for our patients.

More recently, we have started to exploit tumor surface antigens as therapeutic targets for an emerging class of therapies such as antibody drug conjugates, or ADCs, and bispecifics. As you can see, lung cancer is really not a single uniform diagnosis where we can apply a one size fits all approach, but it really spans multiple subtypes and requires a personalized approach.

Each of these layers of classification brings additional diagnostic insights, new predictive biomarkers and ultimately therapeutic opportunities that we can continue to refine precision medicine for our patients with non-small cell lung cancer.

Progress of Systematic Therapies in NSCLC

Here, again, this is to highlight the progress of systemic therapies that have been approved for patients with advanced and metastatic non-small cell lung cancer in the last 20 years. What is really exciting to see is that we have been able to translate some of these advances and breakthroughs into the early-stage setting as well. We have shown that by bringing some of these really powerful treatments into the earlier stages, we have been able to improve outcomes in this space where there is potential for a cure, right?

Dr. Paz-Ares and Dr. Chaft will review some of these trials and clinical evidence in greater detail.

Immunotherapy Approvals for Neoadjuvant and Adjuvant Therapies in Early-Stage NSCLC

One major area of transformation has been really the development and the approval of immune checkpoint inhibitors and IO as neoadjuvant and adjuvant therapies in early-stage non-small cell lung cancer, where now we have several regimens of immune checkpoint inhibitors in the neoadjuvant, perioperative and adjuvant setting that are available and now really standard of care for our patients with early-stage non-small cell lung cancer.

Landscape of Actionable Biomarkers in NSCLC (2026)

Here is the list of actionable biomarkers in non-small cell lung cancer that are tied to specific therapy approvals and therefore really required for clinical testing. On the top, these are the biomarkers for advanced and metastatic stage tumors with molecular targets on the left, including things like EGFR mutations, ALK fusions, ROS1 fusions and many others.

Then protein IHC markers on the right. These are markers that many of us are quite familiar with, right? Now we also have biomarkers that are required for testing in the early-stage setting as well, which are being used to guide different neoadjuvant and adjuvant therapies. This includes EGFR classical mutations, ALK fusions and PD-L1 IHC.

For this seminar, we are really going to focus on the biomarkers in early-stage non-small cell lung cancer and discuss their clinical role and how to really improve biomarker testing for early-stage, which is an area of major unmet clinical need.

Here, there are really three major ways that biomarkers drive clinical care for patients with early-stage non-small cell lung cancer. Here we are really talking about stage IB to IIIA and IIIB stage, which roughly translates to tumors greater than three to four centimeters or with node-positivity.

First, these biomarkers are required as part of standard of care treatments based on some of these regulatory approvals. These are primarily classical EGFR mutations, ALK fusions and PD-L1. If it is positive for EGFR or ALK, then neoadjuvant IO is contraindicated. Although not yet FDA approved, there is an option for neoadjuvant TKI for EGFR-positive tumors. Then after surgery, adjuvant TKI for EGFR and ALK are recommended rather than adjuvant IO.

Now if they are negative for EGFR and ALK, then neoadjuvant chemo IO can be given with PD-L1 IHC though not required, can actually provide some predictive information with higher expression being linked to greater benefit. Then, of course, after surgery, adjuvant IO can be given with PD-L1 expression being required in select scenarios.

Then next there is an emerging clinical practice to really tailor perioperative IO based on some of these genomic biomarkers. Because we know that while IO is very effective for some patients, it does not work for all patients and their side effects are really not trivial. We know that patients with ROS1, RET, NTRK and ERBB2 are quite similar to EGFR and ALK, and that they are less associated with smoking exposure and perhaps have a less immunogenic tumors that are less likely to benefit from IO.

In addition, prior IO exposure really increases the risk of toxicity with subsequent targeted therapy. Based on these clinical evidence, the field is really moving towards neoadjuvant and adjuvant chemo without IO in these patients where IO is less likely to be effective and come with increased toxicity risks.

Lastly, there are numerous clinical trials that are exploring different targeted therapy options in the neoadjuvant and adjuvant space for many targets, like KRAS G12C, RET, ROS1, ERBB2 mutations and EGFR exon 20 and PACC mutations, really similar to what we did for EGFR and ALK.

These trials are an integral part of patient care today and really critical for moving the field forward. So early-stage biomarker testing is really the foundation for future progress.

Biomarkers for Perioperative Treatment in Early-Stage NSCLC

Given this clinical workflow, what do the guidelines say about what is recommended for testing in early-stage non-small cell lung cancer?

Here, classical EGFR mutations, ALK fusions and PD-L1/IHC are recommended for testing at minimum by several guidelines, including the NCCN, ESMO and the IASLC consensus, given the FDA and EMA approvals for these perioperative therapies with specific biomarker indications.

Then, as mentioned, we have several other biomarkers that even though they lack specific approvals, they are quite useful for personalizing perioperative IO and for clinical trial enrolment. These include things like RET fusions, ERBB2 mutations and many others. Their role in clinical decision-making is currently being increasingly recognized by the NCCN and several other consensus statements. Given this landscape of three must-test biomarkers and this ever growing list of emerging other biomarkers that drive clinical care in trials, a comprehensive biomarker testing approach using NGS or next-generation sequencing and PD-L1 is really the best practice for early-stage non-small cell lung cancer and likely the most economical and tissue-efficient approach, similar to what we have learned from our metastatic setting.

Again, this is also being increasingly recommended and supported by guidelines and expert consensus statements. Dr. Lopez-Rios will go into greater detail regarding these different diagnostic assays from a more laboratory and technical perspective.

Logistical Considerations for Tissue-Based Testing

Here, we discussed how comprehensive biomarker testing is really important for patients with early-stage non-small cell lung cancer. Now let us go behind the scenes in the lab and talk about how we can really set ourselves up for success.

First, it really starts with the specimen collection, where we want to ensure that we have adequate material for biomarker testing. We could do this by incorporating Rapid On-Site Evaluation, or ROSE, during the actual procedure. Then we prepare the material in the lab using appropriate protocols. Then we, as pathologists, have to make sure that we are judicious with our use of diagnostic IHC stains during the workup and really prioritize the material and think about biomarker testing when we are doing these analysis.

Then, of course, we go ahead and perform comprehensive biomarker testing using IHC, PCR and NGS as needed, ideally through a pathology initiated reflex testing protocol. Once the results come out, then we have to report the biomarker findings in a way that clearly annotates actionability through a multidisciplinary framework in mind and also in a timely fashion. We will talk more about why turnaround time is such an important issue for early-stage.

As you can see, this is easier said than done. This is quite a bit of a multi-step process where we can encounter some challenges. Let us talk about some of these problematic issues and also some potential solutions that we can come up with in this space.

Challenges and Solutions for Testing in Early-Stage NSCLC

Biomarkers for early-stage non-small cell lung cancer is really significantly under-tested. There is a US study suggesting rates as low as 50% in eligible cases. We really need to increase awareness, continue to promote education as we are doing now today, and then incorporate testing protocols like reflex testing to hopefully increase uptake of testing.

One critical factor, especially for neoadjuvant treatments, is really the need for ultra rapid turnaround time. That is because there is a narrow window of resectability that can be jeopardized. We really need to finalize all the different types of neoadjuvant treatment decisions as soon as possible.

Thankfully, we now have some ultra rapid NGS solutions that can return the result in a matter of days rather than weeks. If that is not available, then I would say PCR/IHC that is more rapid can be considered in select scenarios. Again, reflex testing can also shorten the turnaround time as well.

Tissue is still the issue here. In about 20% to 30% of cases, I would say we get insufficient biopsies for testing. For this, we can consider ROSE and reflex testing again to optimize and streamline tissue stewardship. While it is not helpful for neoadjuvant, testing on the surgical resections could be a helpful backup as well as PCR and IHC in cases, where, let us say, NGS is not feasible in these scan biopsies.

One thing that I wanted to highlight here is that ctDNA-based liquid biopsy for genotyping is not really recommended for early-stage in contrast to metastatic stage. That is really due to issues related to sensitivity in these small early stage tumors.

Then another issue that can be quite a bit of a challenge is that staging information is not always available when we, as pathologists, are considering to send a biopsy for molecular testing. This is really important since testing is not necessary for very small stage tumors, so I think we need to continue optimizing multidisciplinary communication and also standardized protocols for clinical staging to help ensure that we are doing stage appropriate testing, which is, again, greater than three to four centimeter or node positivity.

Lastly, there can be some issues related to suboptimal reimbursement and some of these issues with limited resources for testing, particularly for broad NGS and early-stage, because this is relatively a new concept, right? Here we need to engage with the payers and insurance early on. We need to continue to build clinical evidence for why NGS is so important and so helpful in the early-stage and then continue advocating for our patients.

Collaborating with biopharma and testing companies can also be helpful. They may be able to assist and support in some aspects of the funding.

Summary of Biomarker Testing in Early-Stage NSCLC

Just to summarize our discussion on biomarker testing. Here, we are talking again on patients with early-stage non-small cell lung cancer. This is really to help personalize both neoadjuvant and adjuvant treatments. Testing should be ideally done at the moment of diagnosis before considering these patients on systemic therapies.

Specifically, we should be testing for EGFR, ALK and PD-L1 at minimum. We really need to consider and really perform an expanded panel of multi genes for other biomarkers, if at all possible, using a broad panel NGS and IHC.

Again, testing for early stages is really critical for some of these three important clinical indications. It is number one to deliver standard of care clinical practice based on these biomarker based approvals. Number two, to help optimize patient selection for perioperative IO to really maximize benefit and then also to minimize toxicity risks.

Lastly, it is to support perioperative targeted therapy trials to deliver these new therapeutic breakthroughs and for the future.

Importantly, we also talked about the value of optimizing tissue workflows and stewardship, the importance and the benefit of pathology initiated biomarker testing workflows, and really embracing the spirit of multidisciplinary collaboration throughout the entire testing journey to really enhance biomarker testing and precision medicine for our patients with early-stage non-small cell lung cancer.

With that, I thank you for your attention. Now I will turn it over to Dr. Fernando Lopez-Rios, where he will continue our discussion of molecular testing and interpretation.

Optimizing Timely Molecular Testing and Interpretation (II)

Dr. Fernando Lopez-Rios (Hospital Universitario 12 de Octubre): Thank you very much. I am delighted to be here today. Very grateful for the invitation. I will, in the next few minutes, also try to emphasize the idea of trying to include all patients with lung cancer in our testing schemes, regardless of stage or even histological subtype.

Contents

This is the content of my presentation. I will start by trying to understand the clinical practice gaps in patients with advanced lung cancer, because there are some lessons that we can learn if we want to make this a reality. Then I am going to try to explore how to transform lung cancer care by including all patients. There, I will focus on turnaround time and also in trying to organize very effective communication strategies.

Begin With the End in Mind: Turnaround Time Is Becoming More Important

Let us look at the clinical practice gaps in patients with advanced lung cancer. There have been many studies trying to look into this situation. Probably a fair summary would be using those four T's. As Dr. Yang has just mentioned, sometimes there is just not enough tissue. We struggle trying to find the right assay for our testing strategies.

Some of our efforts sometimes take too long, so turnaround is a huge problem. Of course some of these assays are not cheap.

I am going to talk today mainly about turnaround time because it is something we can somewhat control. Also, it is very interesting because if we manage to include all patients, then we are going to have better turnaround times for our patients with advanced lung cancer, so that is going to be a very welcome benefit of this universal strategy.

Begin With the End in Mind: Turnaround Time Is Becoming More Important

The reason why turnaround time is a huge problem, a huge practice gap is because as you can see there in that prospective observational study that has just been published because physicians cannot wait and patients cannot wait. That was the major reason why biomarker test results were not available before starting first-line therapy.

What underlines this statement is that next-generation sequencing NGS just takes too long. As you can see there, the median NGS turnaround time in this study was close to one month for tissue NGS and it was a lot better for liquid biopsies, but still more than 10 days, around 12 days.

Begin With the End in Mind: Biomarker Test Results Available Before Starting 1L Therapy Associated With Improved OS

We need to focus in turnaround time because, as you know, the overall survival of patients with advanced lung cancer is much better if the results are available before the beginning of first-line therapy. That is something that we all need to understand and probably prioritize when we organize our clinical and biomarker testing workflows.

Begin With the End in Mind: How Long Can My Patients Wait for Results?

If we try to address this in a comprehensive manner, including all patients, regardless of histology or stage, it is nice to try to remember this law of management, which essentially can be summarized by a very simple idiom, just keep it simple.

When we try to organize our testing workflows by selecting patients according to stage or histological subtype, and I think Dr. Yang might agree with me on this. The workflow sometimes looks like that tangle line there at the bottom of the slide. It is not that easy because of the ways some of these assays are organized, you need a certain number of patients to fill in those NGS chips.

If you try to include all patients regardless of stage, then your clinical and molecular workflow, it is much leaner. This is what we are trying to do. That is what I will try to address in the next few minutes.

How Can We Transform Lung Cancer Care? Exploring Whether Testing All Patients Is Feasible

This is how we envisioned the journey of a patient with lung cancer from a biomarker perspective. This is what we are now trying to do in all patients with lung cancer, regardless of histology or stage.

The most important thing is to organize a reflex workflow, which we have previously concentrated with the whole clinical team and particularly with oncology. Then I am going to focus today on those three topics there. The first one would be the need to make sure that we balance diagnostic immunohistochemistry or the need to be very accurate when we diagnose our patients with lung cancer with the future use of tissue for biomarker testing. That is a very difficult thing to achieve.

I am also going to very briefly try to convince you that next-generation sequencing is really the only way to make sure that we do not miss any actionable target in our patients with lung cancer.

Finally, I will very briefly mention some effective communication strategies. What I am trying to say here is that this is so complicated with so much information that it is a very good idea to try to set up several layers of communication to make sure that you do not miss any relevant information when drafting and discussing these molecular reports.

Histologic Diagnosis: Sensible Use of Diagnostic Immunohistochemistry

Let us just start with the issue of trying to not exhaust the very little tissue that sometimes we receive when we try to classify tumors. I have to say that I am many times very frustrated with my surgical pathology colleagues or other pathologies in other institutions because I personally think that we order too much diagnostic immunohistochemistry. The result is that sometimes there is just not enough tissue for IHC-based biomarkers or for NGS.

Currently, we are only ordering TTF1 and p40 if we do not see glands which would qualify as an adenocarcinoma or keratin which would be an squamous cell carcinoma. Many times, we really do not know if we are in front of an adenocarcinoma or a squamous cell carcinoma, but we try to stop there and preserve the tissue for further testing.

Then we also try not to order neuroendocrine markers unless we see very clear neuroendocrine morphology. I know this is easier said than done, but we try to be very conservative in the use of diagnostic immunohistochemistry.

Histologic Diagnosis: Sensible Use of Diagnostic Immunohistochemistry

There are several guidelines out there that have tried to emphasize these. We all need to be aware that sometimes it is better to look at the clinical history before ordering more immunohistochemistry, because sometimes the answer is just right there and you do not need to use more of that tissue to look for those answers.

Exploring the Clinical Utility of Computational Pathology: Digital PD-L1 Scoring is Faster and More Sensitive

Something else. We are doing that saves a little time is the use of artificial intelligence algorithms to score PD-L1, for example, or tumor content. This is really faster than pathologists, as you can see there in green. Not only it is faster, but it is also probably more sensitive for the identification of lower cut-offs or for the scoring of tumor content in difficult cases. I am a huge advocate of some of these algorithms that can streamline the whole lung cancer testing workflow.

Can We Start With Single-Gene Assays? What Are the Pros and Cons of Different Methods (or Assays)?

As I said before, when we organize our testing strategy, it is important to consider those parameters there in the table. These refers to EGFR mutation testing, but it can be easily extrapolated to other predictive biomarkers. We need to understand the analytical sensitivity of our technology. The breadth, the width, which sometimes is called diagnostic sensitivity of our assay. Of course, we need to really know if we are able to precisely annotate variants, for example, the input DNA or RNA that we are going to need.

Of course, the cost and for sure, the turnaround time. What I am trying to say here is that if we use all this information, probably the only way to be very comprehensive when we try to look even for EGFR mutations or ALK fusions is by using next-generation sequencing. So probably with the arrival of other predictive biomarkers in the early-stage setting, there will be no other way to really address this challenge also in our patients with early-stage lung cancer.

Implementing Effective Communication Strategies: Reclaiming Conversation

In the final few slides, I would also like to emphasize the importance of very effective communication strategies, because again, now there is so much information that it is really easy sometimes to not only miss things, but also not be that fast when you are trying to release results.

The first thing that I like to share with you is probably the need of daily operational meetings in in pathology and molecular pathology labs. What I am trying here to do is to just reclaim conversation. These are very simple stand up meetings between molecular pathologists and technicians that can really streamline the molecular workflow, particularly for NGS because you need a certain number of patients to start your NGS procedure. This really simplifies the decision-making process.

Implementing Effective Communication Strategies: Use of Checklists

Another interesting and underutilized tool in medicine is the use of checklists. This is an example for patients with non-small cell lung cancer. Essentially you just very briefly and very quickly review the clinical information, the biomarker information and you just go over all the clinical, pathological and biomarker test results, trying to make sure that you do not miss anything when you draft your molecular pathology reports, which I think they should be really easy to interpret.

Implementing Effective Communication Strategies: Easy-to-Interpret Molecular Pathology Reports

They should be structured in sections. Some of them are very obvious and, of course, we are well aware of the initial sections that some of these reports need to contain.

For example, a few things might be worth mentioning. For many times I am looking at pathology reports and molecular pathology reports that they do not explicitly incorporate the information of not only of the specimen in which the method was carried out, but the exact paraffin block that was sent for molecular testing, for example.

As also Dr. Yang has mentioned, the results would be structured using a standardized nomenclature. We need to look into the pathogenicity and also the actionability of the results. It is also very important to make sure that we correlate the results with previous testing in other specimens in other surgical events. Also if other assays beyond NGS are ordered for orthogonal testing, it is also relevant to try to include this information in the same report.

Implementing Effective Communication Strategies: Apply Common Sense When Looking for Information

These are some of the sources that we are using to understand the reports and draft these reports. I also have to say that the field is moving so fast that sometimes you really need to use a PubMed or even Google to try to come up with common sense when making some of the final decisions of what to incorporate in some of these reports.

Implementing Effective Communication Strategies: Molecular Tumor Boards

The last layer of complexity in effective communication would be the integration and discussion of results at molecular tumor boards.

Conclusion: Remember Parkinson’s Law

With all this in mind, I wanted to finally share with you this idea of maybe challenging the recommended turnaround time for patients with non-small cell lung cancer using next-generation sequencing, but trying to incorporate patients with earlier stages, so then we might have at the end, a wonderful situation where all patients with lung cancer have ultra-fast results in maybe just a few days’ time. So I like this Parkinson's law of management, I think comes in handy for this final idea.

Thank you very much for your attention. Now, over to you, Ryan. Thank you.

Let's Return to a Few Questions From Earlier

Speaker: Thank you, Dr. Yang and Dr. Lopez-Rios for a wonderful discussion of best practices in testing in this space. Now we will briefly return to some of the questions from earlier in the program before our panel discusses this topic in more depth.

Posttest 1

Let us return to our first pre-test question. At minimum, which of the following should be assessed at diagnosis of early-stage non-small cell lung cancer to inform approved perioperative treatment options. Please review the answer options and vote now.

Let us close the poll, please. Okay. Excellent. Very well done. I will move on to the next slide.

Posttest 1

Of course, our answer is EGFR, ALK and PD-L1.

Posttest 2

Our next posttest question, is 64-year-old patient with newly diagnosed stage IIIA resectable non-small cell lung cancer, undergoes comprehensive biomarker testing. Results reveal an ALK fusion and PD-L1 tumor proportion score of 60%. Based on these findings, along with available evidence and guideline-informed clinical practice, which of the following statements best reflects the role of immune checkpoint inhibitor therapy in the perioperative setting for this patient. Please review those answer options and vote now.

Let us close the poll, please.

Posttest 2: Rationale

Okay. Thank you for voting. Our optimal answer for this one is ALK positive non-small cell lung cancer may derive limited benefit from any perioperative ICI therapy despite elevated PD-L1 expression.

Posttest 3

Our last post-test question before we move to our flipped roundtable. You are discussing comprehensive molecular testing for patients with early-stage non-small cell lung cancer with your team. During your discussion, you tell them, all of the following are reasons to use comprehensive molecular testing in this setting, except which one? Please review the options on your screen and vote now.

Let us close the poll if we could. Okay. Very good. What we were looking for here was reduced insufficient tissue barriers through ctDNA-based liquid biopsy genotyping.

“Flipped” Roundtable What Would You Like the Experts to Discuss?

Thank you very much for your responses. We will now move to our flipped roundtable discussion, in which our entire panel of experts will discuss the topics that you are most interested in.

Flipped Roundtable: Poll 3

Which of the following questions would you like the experts to discuss further regarding molecular testing in early-stage non-small cell lung cancer?

Experts are waiting with bated breath and lets please close the poll.

Okay. Dr. Yang, we have our answer. If you would like, please take it away.

Dr. Yang: Yes, for sure. Lots of really important topics here. All of us, or at least the majority of us here chose option B, which is strategies to ensure that our patients receive NGS in a timely manner. Here we are thinking about whether to do this in a reflex way or an on-demand testing.

I guess I will get started with this issue. As we mentioned in our discussion today, I really do think that turnaround time is really critical when we are thinking about incorporating these testing strategies for early stage, especially for really the neoadjuvant treatments. That is again, because we have this window of susceptibility that we really need to think about. Because we are already considering about delaying surgery and delaying treatment for neoadjuvant treatment, we really need to prioritize and expedite testing to make sure that we have all the information that is available.

What that means to us, as pathologists, is how can we figure out a better way to improve the turnaround time? I really do think that reflex testing is probably one of the more important and more simple workflow issue that we can implement. Maybe not so simple, but at least, it is not as expensive as bringing new technologies on board.

The reason why pathology initiated reflex testing is so helpful is because when we are evaluating the sample, we will be thinking about biomarker testing and we will send off all the necessary assays that are needed at the moment of diagnosis without having to wait for any other information regarding the requests from other clinical colleagues.

By the time we get a request from a clinical colleague, it is probably too late. Having a system in place where we, as pathologists, are in the driver's seat and making these important decisions about sending testing can really help with the turnaround time.

As mentioned, we do have several assays that are now half shorter runtime, especially on the NGS base. There is a lot of ultra-rapid NGS solutions that can give us answers in days rather than weeks, which has been really helpful for this faster turnaround time.

Also thinking about using IHC and PCR in select scenarios, which may be quicker than some of the more conventional NGS, which can take more time. Here, I am also curious to get inputs and thoughts from Fernando, regarding some of the issues here, regarding testing for early-stage.

Dr. Lopez-Rios: Do you want me to start then? Okay. It is more of a cultural thing. We need to maybe convince everyone that, including all patients, really makes everyone's life easier because then the NGS will be a lot cheaper. Then you do not need to struggle organizing and selecting and deciding who goes where in terms of different testing pathways. For me, it is more a mental thing that we tend to say, okay, so we only need to look for three now actionable biomarkers may be a fourth round the corner. Maybe it is not enough to justify NGS in earlier stages.

If you think of the potential benefit for patients with advanced lung cancer that are going to get an ultra-fast NGS result, then it is probably the tipping point for convincing everyone.

In our institution, right, Luiz, we now have these consensus way to reflex protocol. I have to say it is fantastic. We order NGS the very moment we diagnose a malignancy, a primary lung tumor in a small biopsy or core needle biopsy even before we have signed out the surgical path report.

Because now all the surgical pathologists are convinced that the sooner you start the procedure, the better for everyone. That is working for us. Maybe your thoughts, Jamie and Luiz. How do you envision this?

Dr. Luis Paz-Ares (Hospital Universitario 12 de Octubre): I mean, if I may start, it is very important, typically, at least for patients from our institutions, the time we discuss them into the MDT, we already have the NGS results. That means before I actually have seen the patient, even for the first time, we discussed the patient that is typically presented by the pulmonologist. Actually, we have all the information on staging, on resectability, but also on the health status of the patient, how good candidates are for surgery and so on.

Also we have all the NGS and all other biomarker information, so that we may decide there if the patient should go on a systemic therapy to start with, and which type of systemic therapy. Truly, it is not only to talk about those genomic tests that we already have direct implications such as EGFR or ALK profiling, but some others are very important.

Let us say if we see a patient that is having a RET fusion maybe is not going to be a very good idea to actually treat this patient with chemo plus immunotherapy even the patient is actually having a high expression of PD-L1 into the tumor, because we know that that may hamper the availability for receiving targeted therapy in the near future. It is very relevant information I would say.

Dr. Jamie Chaft (Memorial Sloan Kettering Cancer Center): I will add the one place where we diverge from this reflexive NGS on everyone, which is still important, is it is not uncommon for us to see a patient who is appropriate for induction therapy coming from the outside, where there just is not any tissue to test. That is a scenario where rebiopsy and efficient communication with your pathologist that says I need EGFR and ALK in a week is essential, and sometimes PCR-based testing might be preferable there because you will get a much faster turnaround, at least for Soo and myself, our NGS panel takes a long time. So we have a rapid NGS in addition to the extensive NGS, and it is quite complex.

Those cases you can always start the chemo and add the immunotherapy when you have confirmed none of the important alterations are there. I will add HER2 and ROS to the mix. It is not just RET. There are many alterations and reasons we do NGS, and I will talk you through some of the studies that are coming for these specific patient populations in the next hour.

Dr. Yang: Yes. There is a lot of learned lessons from the metastatic in advanced stage setting that we are able to apply. I really like the concepts that, Fernando, that you discussed that actually doing NGS upfront for all of these early-stage non-small cell lung cancer patients actually really simplifies our workflow in the pathology lab. We get the most amount of information, useful and helpful information as much as possible from these small biopsies.

I do not know if we have time, but I guess the other thing that we can think about that I just mentioned really briefly is this potential use of liquid biopsy for early-stage non-small cell lung cancer. Here I want to just make a distinction, and this is a liquid biopsy for genotyping that we use for metastatic lung cancer. It is not really for MRD or for cancer screening. Again, for these type of conventional genotyping liquid biopsy platforms, why is this not helpful for early-stage non-small cell lung cancer.

Dr. Chaft: Just not sensitive enough. We have got some good data out of the Lung Cancer Mutation Consortium project, really for stage I to III, we are seeing a detection rate in the 20% to 30% range. Certainly better for squamous where we are not likely to see the actionable drivers anyway, but definitely to be avoided in the absence of perhaps stage IIIC, where you are not planning surgery.

Dr. Lopez-Rios: Yes, fully agree. We are not quite there yet, and we need to balance sensitivity versus specificity and not easy to do, particularly with larger panels, which are the ones that we are probably going to need in the space. I am optimistic, but I am cautious. We still need a bit more time to make this a reality.

Dr. Paz-Ares: Yes, I agree. We have still some room for improvement here. It is a nice technology but not sure it is ready for prime time as yet.

Applying Molecular Testing Results to Perioperative Systemic Therapy Selection

Dr. Yang: With that, now we will move on to the next portion of our talk by Dr. Luiz Paz-Ares, Applying Molecular Testing Results to Perioperative Systemic Therapy Selection.

Dr. Paz-Ares: Thank you very much, Ryum. Thank you, Fernando, as well. Both of you had really very well set up the stage for why should we have the information on the biomarkers in this context of early-stage non-small cell lung cancer.

Case: Patient With Early-Stage NSCLC

Before starting reviewing some of the available data, I am just going to comment this case of an early-stage non-small cell lung cancer. A 79-year-old woman that presented with coughing and the patient is really a long-term smoker. CT scan showed a left upper lobe mass of about a bit more than five centimeters with significant locoregional adenopathy. The PET was positive in the tumor but also in the lymph nodes.

The biopsy showed an adenocarcinoma of the lung being the mediastinal lymph nodes negatives. The tumor was finally staged as T3 by the size of the tumor N1M0, stage IIIA. The biomarker testing showed a wild-type EGFR and ALK, and immunohistochemistry showed PD-L1 expression in 95% of the cells.

The treatment plan was neoadjuvant or perioperative therapy.

Poll 4

Based on that, the question here is going to be, in your current practice, which of the following would you consider to the optimal treatment of this patient?

1. No preoperative therapy; adjuvant chemotherapy alone;
2. Preoperative chemotherapy alone, followed by adjuvant chemo;
3. Preoperative immunotherapy alone followed by adjuvant immunotherapy; and
4. Finally, preoperative chemotherapy plus immunotherapy followed by adjuvant immunotherapy.

Please vote. I suppose that we are done, right? In that case, I am just going to maybe move to the next session and then we will make comment on the result of this question.

Immunotherapy in Early-Stage NSCLC

Immunotherapy Approvals for Neoadjuvant and Adjuvant Therapies in Early-Stage NSCLC

What I am going to do is I am going to review the available data in terms of clinical trials supporting the use of immunotherapy treatments in the early stage. Then we will also review the data on targeted therapies.

In terms of the immunotherapy, we have data supporting the use of some of these PD-1 or PD-L1 inhibitors in the neoadjuvant context, but also as perioperative treatment and also as adjuvant treatment alone. For that, we have data with nivolumab combined with chemo in the neoadjuvant setting. We have data for durvalumab, pembrolizumab and nivolumab in the perioperative setting, meaning by that induction treatment before surgery, then surgery and finally in combination with chemotherapy and then after surgery immunotherapy alone with all these three agents.

Finally, we have some data. Actually the initial data in the adjuvant setting with atezolizumab and pembrolizumab. We will start with those.

Phase III Adjuvant Immunotherapy Trials

Here you see the typical design of these two trials, the IMpower010 and the PEARLS trial with pembrolizumab on the right and atezolizumab on the left. Patients were typically stage IB to III. They were treated in the IMpower with adjuvant chemotherapy and then randomized to receive after four courses of chemo atezolizumab versus best supportive care.

In the PEARLS trial, the same type of patients were randomized to pembrolizumab or placebo. Important to say here, in the PEARLS trial, some 15% of the patients did not receive chemotherapy. They denied chemo or were not eligible for.

The primary endpoint was DFS, but the patient population were somehow different. There were three patient population in the atezolizumab trial stage II to IIIA with positive staining of PD-L1 or any patient with stage I, II to IIIA or the ITT population.

For the pembrolizumab, the two populations were the overall population and the PD-L1 more than 50% staining tumors.

In any case, the primary endpoint was DFS in the two trials.

Adjuvant IO Trials: DFS

You see here very consistent results, clearly improvement on DFS with a hazard ratio in the range of 0.7 to 0.76. That means a decrease in the progression rate all along the study period in the range of 25% to 30%, very consistent data. Where the data were not, I would say, consistent, whereas in the influence of PD-L1 expression in the atezolizumab trial, those patients with high expression were doing better. That tendency was not that clear in the PEARLS trial. Important to say, the PEARLS trial data on overall survival is still pending.

CheckMate 816: Neoadjuvant Nivolumab + CT for Resectable Stage IB-IIIA NSCLC

The neoadjuvant data were available for chemotherapy plus nivolumab. That trial was actually performed the CheckMate 816 on patients with stage IB to III with EGFR and ALK-negative disease. They were randomized to receive chemo alone control arm for three courses versus chemo IO with nivolumab three courses as well. Then patients were operated on and then optional adjuvant treatment with chemotherapy or radiation was offered.

CheckMate 816: 4-Yr OS

You see here that those patients treated with chemo IO with chemo plus nivolumab were doing better with a clear improvement in OS, with a hazard ratio in the range of 0.72. That means a decrease in the range of the risk of death all along the study period of about 28%. That translates into an improvement in overall survival at five years of 10%, between 55% to 65%.

Important to say it, EFS was also positive, with a hazard ratio of 0.68. There was a clear improvement in the pathological complete response rate.

Ongoing Phase III Perioperative Studies in Resectable NSCLC

Very consistent data were actually seen in the perioperative trials. There were a number of those trials with very similar design. You see here, the typical design stage II to III patients with resectable disease and good PS and typically for most of the cases ALK and EGFR wild-type. The only exception were the KEYNOTE-671, where patients with ALK or EGFR were allowed.

Those patients were receiving four courses of chemo IO versus chemo plus placebo, then operated on and finally adjuvant phase, patients were receiving immunotherapy alone post-surgery for about 12 to 16 cycles or placebo, depending on the treatment arm.

Phase III Perioperative Studies in Resectable NSCLC: pCR and EFS

You see here that the results were very consistent all along those three trials with durvalumab, nivolumab or pembrolizumab. You see here, hazard ratios are very consistent actually in the range of 0.6 overall and that relating to a clear improvement as well in the proportion of pathological CR rates, I would say in the range of 2% to 5% in the control arm versus a 17% to 25%, something like five to seven times higher in the chemoimmunotherapy arm.

You see here, as we have said, a clear improvement in event-free survival. That was seen not only in those patients with stage III, but also in stage II. The data were very consistent because all the benefit was seen across the different levels of PD-L1 for most of the cases, even the benefit was maybe better was higher the magnitude of benefit for those patients with high expression as compared to the PD-L1 negatives, where some benefit we are seeing there, but not of the same magnitude.

KEYNOTE-671: 5-Yr OS

You see here the data of the KEYNOTE-671, where the data are very mature for overall survival. You see here again a survival benefit at five years of about 10%, 54% versus 65% at five years, very similar to what we have seen with the chemotherapy plus nivolumab in the neoadjuvant trial.

Targeted Therapy in Early-Stage NSCLC

Trials Targeting Common EGFR Mutations (Ex19Del or L858R) in Early-Stage/Locally Advanced Disease

In terms of the targeted therapy, we have particularly three relevant subsets of data, two in the adjuvant setting with osimertinib for early-stage after surgery or with osimertinib alone for those patients treated post-chemo radiation with stage III. We have also some data with the NeoADAURA, which is in the neoadjuvant setting, chemotherapy plus osimertinib in patients that are in stage II to IIIB with resectable disease.

Of course, all of that for patients with EGFR-mutant disease, classical mutations.

ADAURA: Adjuvant Osimertinib After Complete Resection for Stage IB-IIIA EGFR-Mutated NSCLC

You go to the next slide, you realize here this is the design of the ADAURA trial. Patients with stage IB to IIIA tied with EGFR exon 19 or 21 mutations were randomized to receive osimertinib versus placebo. Those patients were randomized after surgery and the duration of treatment was for three years.

ADAURA: Results in Overall Population

As you see here, there was a clear improvement in disease-free survival with a hazard ratio of 0.28. That means you are decreasing the risk of progression in this study in about 75%. That means you are decreasing the risk of relapse in three patients out of four that are progressing in the control arm.

Important to say after three years that you are stopping treatment. Maybe the risk of relapse is increasing here. That seems to translate into an improvement in overall survival as you see on the curve on the right.

ADAURA: DFS by Subgroup (Overall Population)

The benefit we are seeing all across the different subsets the same. The magnitude of the benefit was very consistent. Maybe those patients with EGFR mutation in exon 19 were somehow doing better as compared to those in the exon 21 with somehow better hazard ratio, 0.24 as compared to 0.45.

LAURA: Osimertinib After Definitive CRT in Unresectable Stage III EGFR-Mutated NSCLC

In the LAURA trial, similar data were seen in the stage III. Those patients were treated with chemo radiation in unresectable disease, and then if they did not progress after chemo radiation were randomized to receive osimertinib versus placebo. Of course, those were patients with a classical mutation, exon 19 or 24, as in the ADAURA trial.

LAURA: Endpoint Results ― PFS and OS

You see here that there was a clear benefit for those patients treated with osimertinib. In that case, osimertinib was given forever indefinitely, and the hazard ratio was 0.24, as you see here. So there was a clear decrease in the risk of death. Median PFS being 39 months as compared to 5.6 months. You are decreasing 85% the risk of relapse in this setting. Of course, the survival data are not very mature. We have to still wait a bit.

ALINA: Adjuvant Alectinib vs Chemotherapy in Patients With Stage IB-IIIA ALK-Positive NSCLC

The final data in the ALK positive setting. Patients that were resected with the stage IB to III were actually randomized post-surgery to receive alectinib for some two years as compared to platinum chemotherapy alone. Important to say in the EGFR trial, In the ADAURA trial, patients were actually treated with chemotherapy before being randomized.

ALINA: Disease-Free Survival (Primary Endpoint)

As you see here, patient receiving alectinib for two years were doing better with improved DFS with a hazard ratio of 0.24. You are again decreasing the risk of relapse for about 75%. That is actually true not only for those patients with stage II to IIIA, but in the overall population as well, in that case with the same hazard ratio.

This is very much what I have to say today. Maybe we can discuss any controversy during the roundtable before.

Back to you, Ryum.

New Directions for Precision Medicine in Early-Stage NSCLC

Dr. Yang: Thank you so much for that presentation. Now we will switch to Dr. Jamie Chaft, where she will be talking about New Directions for Precision Medicine in Early-Stage Non-Small Cell Lung Cancer.

IO Is Not Everything!

Dr. Chaft: Thanks, Soo. It is really just the bottom line in all of lung cancer care is IO is not everything and it is not for everyone. It is really easy just to prescribe preoperative chemo/IO or give adjuvant chemo and then IO. So many patients with tumors with specific actionable oncogenes do not benefit.

Adjuvant TKIs for Resected Stage IB-III NSCLC

Luiz just showed you the data that really sells why it is important and not only important, it is essential for us to identify these patients because not only do these patients not benefit from immunotherapy, if we do not know about their actionable EGFR or ALK alteration, we cannot prescribe these life-changing therapies.

Luiz showed you the consistent data but not the inconsistent data. In the adjuvant setting, there are also negative immunotherapy trials.

Not only does not everyone benefit, we would not be giving these therapies that reduce the risk of recurrence by 70% to 75%, which is truly remarkable.

What About Stage I NSCLC?

What about earlier stage disease? There are results anticipated in the next few years looking at even earlier stage EGFR-driven lung cancers, where we know these drugs are well tolerated and highly effective. How can we identify the patients who need additional therapy even in the earliest stages of lung cancer. These are really the future directions and why Fernando's one size fits all molecular testing approach really does should be extrapolated.

Case: Patient With EGFR+ Early-Stage NSCLC

I was asked to present a case. This is a patient of mine, a 60-year-old woman, never smoker, who presented with cough. She had a T3 multi-station N2 adenocarcinoma with an EGFR exon 19 deletion. She was offered participation in the NeoADAURA study which we will soon look at. The study required a core biopsy, and that is not something she had had. She had been diagnosed by EBUS.

Case: Patient With EGFR+ cT3N2 NSCLC

She refused chemotherapy and was treated off-label admittedly, with neoadjuvant osimertinib with truly a remarkable response. If you can see the shadow of a tumor there of what once was a T3, very large lobulated primary tumor.

Case: Patient With EGFR+ cT3N2 NSCLC

She underwent resection outside and for whatever reason they did a frozen section of a lymph node and said it was positive. Then the surgeon wedged out the primary. It actually happened to be a false positive frozen and upon final pathology was negative. She returned to our clinic and underwent restaging scans. You can see that some uptick in the mediastinum are those nodes or is it reactive?

She had a complete resection done with no residual disease. She would have been a major pathologic response less than or equal to 10% viable tumor cells. A pathologic stage IA tumor, tolerated just one cycle of adjuvant chemotherapy and has since been on adjuvant osimertinib.

NeoADAURA: Neoadjuvant Osimertinib ± CT vs CT Alone in Resectable EGFRm NSCLC

Just a case from my clinic to talk about what we did, and now we have some data to suggest perhaps that is what we should be doing.

This is the three-arm study of NeoADAURA. Patients with EGFR exon 19 or L858R randomized to either chemotherapy and placebo, chemotherapy and osimertinib or osimertinib monotherapy and open-label arm. Standard postoperative chemotherapy was permitted.

As the ADAURA results read out early, the study was amended to provide access to adjuvant osimertinib for the entire study population.

NeoADAURA: Surgery Summary

Here are some numbers. We did not look at these in the neoadjuvant chemo/IO studies, but in the neoadjuvant chemo/IO studies, about 80% of patients make it to surgery and R0 resection rate is even less.

When you have a biomarker matched therapy, nearly all of the patients made it to surgery, 92% and 97% in the osimertinib containing arms with very high R0 resection rates. More importantly, is that time to or ability to prescribe the adjuvant component of this. That is 91% of patients irrespective of treatment arm received sponsored supply adjuvant osimertinib.

NeoADAURA: Depth of Pathologic Response

Here is the primary study endpoint. We knew when designing this study that we would not see complete pathologic responses. Interestingly, the neoadjuvant chemoimmunotherapy studies were all initially designed with MPR, major pathologic response as well until we learned that immunotherapy-induced PCR and many of them were amended.

The data are what they are. We saw only a 2% MPR rate, with chemotherapy substantially better with either osimertinib monotherapy or osimertinib and chemotherapy.

Targeted Therapy Unmet Needs

We, in the US have the NCCN, where we can give preoperative EGFR inhibitors. This is not the case globally. However, despite having adjuvant osimertinib, adjuvant alectinib, adjuvant ensartinib is another phase III study that is positive with ALK, and neoadjuvant osimertinib, we have a huge unmet need.

This is a graph based on old data at this point from Dr. Yang in my institution looking at the incidence of oncogenes using our MSK-IMPACT data. We really find actionable oncogenes in nearly two thirds of our patients. If you estimate that osimertinib is about a quarter, ALK maybe 3%, there is still 40%. These are generous numbers. It is probably a little higher with an actionable oncogene where we do not have a targeted therapy.

What are we going to do about this?

The Future of Targeted Therapy in Resectable NSCLC

There are many studies ongoing, and the LIBRETTO study of selpercatinib, which Dr. Paz-Ares referenced earlier, will be presented next week at ASCO I believe. All of these other studies are ongoing, where we anticipate in the next many years to see if adjuvant targeted therapies in these various populations can improve clinical outcomes.

I will walk you very briefly through the schema so we can save some time for discussion.

Olomorasib for KRAS G12C–Mutated Early NSCLC

Pay attention to this one. We had a question on it. The Olomorasib study is looking both at post chemo radiotherapy as they did in the LAURA study or post-operatively irrespective of whether the patient received neoadjuvant therapy or not as long as there is no pathologic complete response and the patient is immunotherapy eligible, the randomization is to olomorasib with pembrolizumab or adjuvant pembrolizumab, which is considered the standard of care.

Same in the consolidation setting with durvalumab.

Selpercatinib: Adjuvant Therapy for RET Fusion+ Early NSCLC

This is the LIBRETTO study looking at adjuvant selpercatinib. This study also allowed patients after definitive intent therapy, whether that was surgery or chemo radiotherapy, randomized to selpercatinib or placebo. We have seen the positive press release anticipating this presentation at ASCO.

Zongertinib: Adjuvant Therapy for HER2+ Early NSCLC

Zongertinib will be studied in the adjuvant setting for HER2+ non-small cell lung cancer. These are patients who do not benefit from immunotherapy in general. However, they are randomized after either neoadjuvant chemo immunotherapy or adjuvant chemotherapy to zongertinib, or investigator's choice standard of care.

Taletrectinib: Adjuvant Therapy for ROS1-Fusion+ Early NSCLC

Taletrectinib, for those rare ROS1-positive patients being studied in a very similar fashion compared to placebo. Another population that may highly express PD-L1 but does not benefit from immunotherapy.

ALCHEMIST Trials: Adjuvant Targeted and Immunotherapy Based on Biomarkers

Then there are the ALCHEMIST studies we have seen, published and presented now. Erlotinib and crizotinib is soon to be published. There is the adjuvant study of nivolumab run by the NCTN, which I will be presenting next week at ASCO. We have studies ongoing really asking the question can we give immunotherapy concurrently with chemotherapy in patients who have had upfront resection with some suggestions from the Spanish group that this may be better.

Perioperative Durvalumab for Early NSCLC

The ADOPT-Lung study is being done throughout Europe looking at the question do we need the adjuvant immunotherapy after neoadjuvant chemoimmunotherapy in the study using durvalumab extrapolating on the AEGEAN experience and really asking a contribution of phase question.

Adjuvant Durvalumab for Early-Stage NSCLC

A little less ambitious, in the US, we are doing the INSIGHT study, where patients with a pathologic complete response are randomized to durvalumab or observation. Very similar question, do we need that adjuvant immunotherapy?

Adjuvant Nivolumab+ CT vs CT for Early-Stage NSCLC

This is the study that I just referenced, the NADIM ADJUVANT study, which is a maturing study, not yet mature but looking very promising. When moving that adjuvant immunotherapy up, all of the studies that Dr. Paz-Ares showed you prescribe immunotherapy after chemotherapy. This is prescribing the adjuvant chemo and immunotherapy together followed by an immunotherapy phase.

NADIM ADJUVANT: DFS

The data are not yet mature but look very good favoring the immunotherapy arm.