A Pathologist Outlines Typical Tumor Tissue Processing Steps after Biopsy in NSCLC
Michelle Shiller, DO, AP/CP, MGP, describes the process of tissue biopsy testing at her institution from start to finish, and Zosia Piotrowska, MD, explains the likelihood of targetable alterations in cases of mNSCLC in patients with no smoking history.
Joel Neal, MD, PhD: Michelle, you probably don't get as many of the clinical details leading up to this case but at what point do you intercept the tissue and then say, here's what we need to do to make sure that when this patient's meeting a medical oncologist, that all the information's ready. How do you shuttle the tissue through these processes?
S. Michelle Shiller, DO, AP/CP, MGP: There's a few options. This would be a situation which the oncologist is ordering the testing. What often will happen, some institutions are very uncomfortable with sending the paraffin block out, so they'll want to send us unstained slides. Locally the unstained slides will be cut and then we also need to see how much tumor distribution there is in the actual block so often they'll stain a slide in H&E stain so then we can assess when we receive the tissue, how much tumor we're working with if we think we actually are even a starting point to be able to do the testing. When we render on the front end that we may be in hot water with respect to tissue quantity sufficiency, we immediately are going back to that pathology group where the tissue originated to ask if there's any other tissue for that patient. One thing that we try to encourage is during a procedure if we get two or three cores to put them in separate blocks so that you can optimize the yield on the tissue because of the three-dimensional orientation of the tumor in the actual core biopsy. If it's in the same procedure, we don't like to do it, but we try sometimes to optimize by going to another block if it's acquired at the same time in the same lesion, not if it's in a different lesion, but if it's in the same lesion. We don't do that, that often. We're actually able. We've fine-tuned the process to where our QNS rate is really quite low. Nonetheless, we try to make that decision quickly and notify everyone because just as Lauren said, sometimes you need to go ahead and move the liquid. Some people do liquid simultaneous to acquiring the tissue biopsy. As Zosia said, there's a lot of different ways to approach all of this, and we're all learning together. Nonetheless, then we first assess it for adequacy, then we have to deparaffinize it and isolate the DNA from the tumor. Then we often inspect photometrically, quantitate the actual DNA and then start the actual next-generation sequencing process. For us, start to finish, that's taking on the average 10 to 11 days from the time it hits the lab to the time we have a result. There's a bioinformatics piece that takes a lot of data that initially comes off and parses it down to something more meaningful, and then it goes through a couple of rounds of PhD providers and then ultimately to one of the molecular pathologists on the team who renders the final interpretation. Sometimes we are calling you guys and asking because we'll see something like compound mutations or something. It's very important in that situation that we understand the patient's clinical phase of therapy. Is it a treatment-naive tumor? What else is going on with this patient so that we can generate a result that's much more meaningful to you guys?
Joel Neal, MD, PhD: You hit on a number of important points of the adequacy of tissue, having a lot of tissue, and then really shepherding that issue through the entire testing process as well as keeping the oncologist, and the treatment team in the loop of what's the turnaround time. By far, my most frequent question to the molecular pathologist is when is this going to be back? Sometimes they say it's going to be back on a certain date and then we'll message them and say, well, something failed and it's going to be an extra week from now because this is tricky. I do like the idea for selected patients doing liquid biopsy so kind of having two things going at the same time. Then I always think about the pretest probability of is this patient likely to have something that I want to wait for a targeted therapy or is this patient somebody who may be a heavy smoker, somebody who has squamous that it's quite unlikely that we're going to use a targeted therapy. Zosia, I know for this patient, never smoking Black woman, adenocarcinoma TTF1 positive. I'd tell trainees who are working with me that the pretest probability of an EGFR mutation was pretty high, like 40%, 50%. How would you reflect on that as well as what's the likelihood of those other actionable alterations roughly?
Zosia Piotrowska, MD: I certainly think especially, given the patient's never-smoking history, I think that in some total, the chances of finding something targetable in a patient like this is probably 50% or even maybe a little bit higher. This is certainly in an enriched population. I think that that is really important when we're making decisions and when we're talking to the patient and explaining I think one of the things that's most important, as you already alluded to, is not just keeping the patient in the dark, but seeing them frequently and really explaining to them why we're waiting for this testing and why it's so very important for treatment decision making. I often also have had the experience that once patients understand that and they understand the idea of personalized medicine and trying to learn as much as we can about the cancer to select the best treatment for them, and then explaining to them that we think that there's a very high chance, in this case, of finding something targetable. Although I will also caution because when I talk to patients, I try not to make it so much like we're really hoping to find something targetable but more to say, we just want to know whether your cancer has one of these mutations or not. I think it can be challenging if you kind of put too much of a kind of push on that oncogene-driven cancer is the thing that we're all hoping for. Even 50% or more patients having an oncogene driver, which means that nearly half don't. If you frame that conversation in such a way that patients can take it, it can be very difficult to deliver the news if they don't have an oncogene driver. Again, this patient also has a positive biomarker in the sense that they have high PD-L1. The way I think about it is really just trying to understand everything we can about this cancer so that when we do finally start treatment, it's the one that we think is best and most likely to help the patient.
