Hi, so in this lecture I'm going to talk about when is a biopsy indicated when you're concerned about lung cancer. I've done two previous lectures, one on screening and one on the probability of cancer determining the probability of cancer when you find a lung nodule. And in this case, what I want to talk about is when do we decide to biopsy something? And this is an important concept because as a pulmonologist, I am often refered patients who need a biopsy. And the truth is most of these patients don't. And we'll talk about how to decide when a biopsy is a useful tool and when it's not the right tool. I like to use the analogy my grandfather gave me, he was a plumber. My grandfather used to say, there is a job for every tool and a tool for every job. And my goal from this lecture is to help you understand when Bob sees the right tool. So the objectives specifically are to understand how the pretest probability of cancer directs the need for a biopsy. And hopefully, you'll be able to recognize the various situations where we do require a biopsy. And for the purpose of this lecture I want to specify that a biopsy. First any non surgical sampling of tissue from the lung node. So unless I say otherwise, if I'm referring to biopsy I'm not talking about surgery, I'm talking about getting a small piece of tissue from the node. And then hopefully you'll understand the patient specific, the physician specific and the technology specific factors that can impact on how and when we decide to do a biopsy. So to start with a question, here is a CAT scan of a small lung nodule of a 68 year old man who presented with a cough. The nodule spiculated, it's about a centimeter and a half in size, and he undergoes a PET scan which shows significant uptake, but only in the nodule. That's the status of the nodule, and now we get some physiologic status on the patient. He's got an FEV1 of 30% of predicted, so he's got very severe emphysema. He's short of breath with just climbing up the one flight of stairs that he has in his house. So what do you do with this lung nodule? Do we send him to surgery to have it removed? Do we attempt to just radiate it, assume it's cancer and just treat it? Do we recommend a follow up CAT scan in six months? Do we biopsy the nodule or do we just say boy, he's just so sick, let's send him to hospice. So, by the end of this lecture we'll hopefully be able to sort out all these options and decide which is best for the 68 year old man. So the outline in this lecture is when we decide to do a biopsy? And what do we do for this pre-test probability of cancer, that we hopefully determine before we get to this point? We're going to talk about people with low probability nodules, we're talk bout intermediate probability, then will talk about those with a high probability cancer. The answer to when we decide the biopsy and how we do it, is different for each of these classes of individuals. And finally, what I want emphasize is what a biopsy can tell you and by association what it can't tell you. So let's talk about this pretest probability. In one of my previous lectures, we talked about how to come up with this probability, and now what do you do with it? So let's just say for argument sake that we're going to call our high probability patients, those with a 70% or more chance of having lung cancer in their nodule. And then we'll have the very low probability people be those with a 5 to 10% probability of cancer, and then everybody else. And as you'll see in a moment that where we precisely put these cutoffs is less important than what we do with the information afterwards. So let's go back to our pendulum, that I referred to in a previous lecture where we talk about the factors that can help us decide what's the chance that there's cancer in this nodule. And there may be more than one nodule in any given patient. But each nodule is going to really have its own probability and what we typically do, is we focus on the highest risk nodule. And as we said earlier risk is determined by these factors. Age, size, what does the borders look like? Is the patient a smoker? Where is it located in the lung? And if you had the PET scan where does it light up? So we then segregate the patient into one of these probabilities. Now high probability nodules are going to found in older larger nodules. Older patients, larger nodules that light up on a PET scan. Low probability nodules are going to be found in younger, smooth nodules, younger patients with perhaps PET negative lesions. But what if you have an old patient with a really suspicious lesion, that doesn't light up on the PET scan? What that's going to do, is put you right into the middle of that intermediate probability. So you can easily envision scenarios that will move that pendulum into one of those three zones for us. Where you set that cut off, you can decide for yourself, and your patient can decide. So the goal with any nodule is to either prove that it's definitely benign or if it they think it's cancer, figure out what it is and how to treat it, that's our ultimate goal. How do we get there? Well again, for those low probability nodules, we could try to prove its benign by taking it out. I don't recommend that for people who have a low probability of cancer. We want to keep as much lung in our patients as we can or we can use time, so take it out in time. On the other hand, if it's a really high probability of cancer, we want to prove it's cancer. Get some tissue under the microscope. Now as I said earlier you can do this with a biopsy or you can just have it resected. And we'll talk about when we decide between those two and how we determine whether those two, which of those two strategies is the appropriate. Now, this middle zone is probably where the biopsy has it's greatest utility. And one way to look at the biopsy is it's just a little bit better than a PET scan at proving you do and don't have cancer. So that intermediate probability is where the biopsy probably has its greatest utility. Remember, I'm referring to a biopsy as a non surgical sampling of the tissue. So when do we do a biopsy? Any time you need to prove it's cancer. The corollary is a biopsy can not prove you don't have cancer. Now I tell this to a lot of patients, and I see a very puzzled look in their face when I see it. And I usually explain it this way, if you look under a microscope, think back to your histology class if you ever took one. And under a microscope, a cancer wasn't all cancer cells, was it? There were some cancer cells and next to that, there may have been some scar tissue or some inflammation, some clot with fibrin or some normal tissue. The point is that even a tumor, a large tumor under a microscope isn't all cancer cells. And when you take a biopsy, you take a small sample of that tissue you might miss the cancer cells. So a biopsy can only prove you do have cancer, it can't prove you don't. And if you repeat that to yourself often enough you will find yourself using biopsies much more appropriately than you otherwise might use them. Absence of evidence is not evidence of absence. So back to this probability line that we've got, and how do we decide where this probability sits and is there a single answer? So I put this together so that you can understand that there are a lot of factors that might affect, what truely counts is a low probably lesion versus what counts is a high probability lesion. For instance, let's just say that the patient is really, really anxious, they can't sleep at night. With the probability that they have cancer, they just can't sleep, it's ruining their life. Well in that situation, you might move this line down a little bit so that somewhere in here is where the biopsy becomes more useful. On the other hand, if you have a safer biopsy technic. Let's say, you have this biopsy tool that some futuristic doctor develops and can safely biopsy the lung with very little risk. Well it becomes a little more justifiable to do more biopsies when the risk is less. But what if we had a blood test that in spite of the appearance of a nodule, could be very reassuring, what we call a high negative predictive value blood test. That if that test is negative, really reassures us that it's not cancer. Well in that case, this line moves way, way up here. And we're going to be much less likely to do invasive procedures if that blood test is negative. If the nodule really really looks benign based on its imaging features. Perhaps it's smooth and round maybe a hint of calcification to it. I'm going to be very loathe to recommend a biopsy to someone that has something that looks very benign. So this line, this cutoff between low probability and intermediate probability is variable where that sits. Let's look at the other end of the spectrum. Let's say, we're deciding between surgery and a biopsy. And this line here is the cutoff between when we decide to just send someone to have surgery or when we decide to do a biopsy. Well, if you have a patient whose surgical risk is really high. Maybe they have pretty bad emphysema, not so bad to scare off the surgeon and the patient, but bad enough that their risk for surgery is higher than average. Well, the burden of proof is going to be quite high. You may want to do a biopsy in this patient to prove to the surgeon and to the patient that they have cancer, and risk of the surgery is worth it. On the other hand, what if you are the best surgeon in the world, and you never had a bad outcome, and all your patients go home two days later and play golf a week after that, regardless of how sick they are? Well if you're that surgeon, then you're probably going to recommend surgery more often than somebody who has a higher complication rate, getting people to recognize this can be hard. So this line is just as variable as this line for different reasons. What we're looking to do is to bring the right answer to the right patient, and that's dependent on the patient, their anxiety level. It's dependent on the nodule itself and it's dependent on the context in where they're being treated. So there's not a single answer to what is the cutoff between low, intermediate, and high probability. It is a fluid answer that depends on where the patient is in that spectrum. So when is it appropriate to biopsy a patient with a lung nodule? I want to emphasize that if you are a patient with a nodule that looks like cancer with a high pre test probability say 80, 90%. And if you're otherwise healthy with good lung function and no cardiac disease, and no other major coorbited illnesses, then me doing a biopsy serves you no use. You probably should go to a surgeon, get that nodule taken out and that's your biopsy, it's also your treatment. Because a negative biopsy in that situation, if you have a pre-test probability that's already very high. A negative biopsy can't move that nodule far enough down the scale for me to be comfortable with saying to you, let's just watch it. In that situation, your surgeon is the one who should be doing the biopsy. Now, if I send the patient to one of my colleagues who is a thoracic surgeon and I work with a lot of very good thoracic surgeons, and they have their conversation with the patient. They say yeah, we think this is cancer and we think we can do an operation to cure this cancer. But you've got some heart disease and you've got lung disease, and I'd like to be 100% sure before I put you to sleep and cut your chest open. I'd like Dr. Arronberg to prove to me that this is cancer before we subject you to the risk of surgery. So the two answers are, we don't biopsy people who are healthy, but if the surgeon thinks they need that answer or the patient demands that answer before surgery,tThat's a situation where it's okay to biopsy highly likely cancer. Keep that in mind. Now, what about other situations where we do a biopsy? Simple answer, when is it appropriate to biopsy a patient with a lung nodule? When we need a diagnosis to facilitate treatment. So a patient who just can't have surgery under any circumstances whatsoever. We think they have cancer, we want to treat it with radiation an excellent way to treat early stage lung cancer particularly for people who can't have surgery. The radiation oncologists, they want to know what they're treating. In that situation it's okay for me to put the patient through the risk of the biopsy, because I'm trying to prove it is lung cancer, I'm not trying to prove it isn't. So when you want to prove it's lung cancer to facilitate treatment, that's a situation where a biopsy is useful. And in this case, the one caveat that I would try to emphasize is that when you biopsy a patient with lung cancer, you want to make sure you're answering not one question, but two. The first obvious one is, is this cancer, but the second one is, what's stage is it? If a patient comes to me with a big mass in their upper or lower lobe of their lung, but they also have big bulky mediastinal lymph nodes or perhaps evidence of metastasis to the adrenal gland on a PET scan. I do them no good by biopsying that lung nodule. We need to prove that that adrenal gland is a metastasis. It gives the patient the benefit of the doubt because the PET scan can be falsely positive. So when I do a biopsy, I don't always biopsy the lung, I biopsy the lymph nodes. Because they're easier and safer, and they tell us about the stage of the disease or I biopsy the metastases for the same reason. Sticking a needle in the lung is a very risky proposition. Sticking a needle in adrenal gland of the liver is less risky and answers the question unequivocally, what stage is this patient's lung cancer? So when we do biopsy it's because it's going to affect treatment decisions, and we want to know exactly what stage disease we're treating. So when is it appropriate to biopsy a patient with a lung nodule, the last. Situation where I think it's reasonable if someone comes to me with a history of another cancer. Particularly a cancer with a high probability of eventually developing matastists. If they have a lung nodule, it might be the matastists, it might be a primary lung cancer. In that situation, regardless of what the treatment options are, we may be willing to do a biopsy to help define the treatment options. So the biopsy is a tool that helps us to find treatment options for the patient. If you don't need that took then skip it. So back to when is a biopsy indicated, the second situation is if biopsy is necessary to make treatment decisions. So we have the sick patient that looks to have lung cancer, but it's early stage. We want to facilitate radiation therapy by giving the radiation oncologist and the patient a clear diagnosis. Even good surgical candidates may require a pre-operative proof. If the surgeon knows that they're going to have to remove an entire lung, to get that lung cancer out. They often will require a biopsy, because that pneumonectomy is a higher risk procedure. We're balancing the risks of the procedure with the risks of the biopsy and when the surgery itself is very high risk. The risk of the biopsy is justified by the information that this is definitely cancer and the surgery's worth it. And then finally, there's this small percentage of patients who after even a good conversation with the surgeon about the risks and benefits of surgery, will say to us, I don't care what you tell me Doc im not doing anything until you prove to me what this is. Thats a very small proportion of patients and should become smaller if we get good at talking with patients about their risks. So let's go back to our 68 year old man, who has a spiculated nodule that shows significant uptake on the PET scan. So he's a smoker In 68, a 15 millimeter nodule with a PET scan, the pretest probability is going to be fairly high. What we then discovered is that this patient has very severe emphysema. He's so short of breath, he can't even climb a flight of stairs without getting out of breath. What do we do about this patient? The answer in this case is to biopsy the lung nod. Now, if he wasn't short of breath, if he didn't have that emphysema, if he was playing golf three days a week. I'd say skip the biopsy, go see the surgeon, get that high pre-test probability nodule removed from your chest. Take home points from this lecture. Don't rush to biopsy every lung nodule. Most of them are benign, but we can't prove it's benign with a biopsy. We can only prove it's benign by taking it out or showing with time. When do we do a biopsy? When we need it for treatment decisions, when the clinical and the imaging clues are discordant and that puts the probability right in the middle. And finally, what a biopsy can tell you, it only can prove you do have cancer, you can not prove the absence of cancer with a biopsy. And as long as you repeat that to yourself enough, you won't misuse this invasive procedure.
