Read the Full Video Transcript

Zach Klaassen: Hi, my name is Dr. Zach Klaassen. I'm a urologic oncologist at the Georgia Cancer Center in Augusta, Georgia. I'm delighted to be joined today by Dr. Alberto Martini, who is a clinical fellow at the MD Anderson Cancer Center. Welcome, Dr. Martini. It's great to have you.

Alberto Martini: Thank you. Thank you very much.

Zach Klaassen: So I'm delighted to have you to chat today about your recent Journal of Urology publication looking at the feasibility of unilateral pelvic lymph node dissection for prostate cancer, specifically in the era of multiparametric MRI and targeted biopsy. So, what was the genesis and thought process behind designing this study?

Alberto Martini: Thank you for the question. So basically, it was, let's say, what I saw mostly in practice while I was in residency. So, I noticed that lymph node invasion contralateral to the, let's say, the prostate lobe with worse disease features was very rare. We all know that there's an important unmet need in identifying those patients that are node-negative, where we can safely avoid lymph node dissection.

Currently, the specificity of the currently available models that we use for selecting candidates for pelvic lymph node dissection are actually hampered by low specificity, and the majority of the times that we remove lymph nodes, we find them to be negative. So, that was actually the trigger that gave me the idea to try this study.


Of course, what we're doing here and the thought process that brought me to this study is actually trying to simplify the tumor behavior. Basically, splitting the prostate in two halves. Of course, there's amplification of tumor behavior, but from a practical standpoint, we only remove lymph nodes on the right side of the pelvis and from the left side of the pelvis.


So my idea was, given that the standard of care has changed, given the fact that now we characterize the index lesion that we know is the driver of also further metastasis as it has been shown in other papers and other research, my hypothesis was basically that one, to see whether we can safely avoid removing lymph nodes only on the ipsilateral side of the pelvis where the index lesion was.


Zach Klaassen: Yeah, absolutely. That's great background. I think your point is well taken, too. Even the nomograms we've used for decades are based before the MRI era as well. So I think-

Alberto Martini: That's correct.


Zach Klaassen: ... that idea, plus the fact that we've got MR, plus the fact that we've got targeted biopsies. So, it's molding that idea into a nice story. So, tell us about the study design. How many centers? How many patients? How did you guys design this study?

Alberto Martini: Yes, and also if I may add something at this point?

Zach Klaassen: Yeah.

Alberto Martini: Also, the earlier studies on lymph node drainage of the prostate were all based on studies where we were characterizing prostate cancer only by DRE-

Zach Klaassen: Right.

Alberto Martini: ... and a systematic biopsy. So, we didn't know where the actual index lesion was. So, that's also something that motivated me into pursuing this study. So the idea was actually, we conceived it in 2021 with a friend of mine who was still in residency, so a few years ago. This guy is now, Massimo Valerio, is a chair in Geneva University Hospital.

We started to collect data. We are part of a network in Europe that is called the Young Academic Urologists, and we are part of the working group on prostate cancer. We wrote a proposal, and we proposed the project to the group. The proposal was very well-received. We got almost 20 centers on board, if I'm not mistaken, in 2019. We collected data from more than 2,300 patients, all diagnosed with mpMRI and targeted biopsy.

We designed a specific data set that allowed us to collect data on a site-specific manner for both the right side and the left side, so that we were able to identify which one was the prostate lobe with worse disease features. We went from there, and we thought we found some pretty interesting results.

Zach Klaassen: Yeah, absolutely. Your group, that's a great idea. A bunch of young urologists in Europe get together, mutual interests, and come together with a nice big sample size, multi-institutional for a project like this. This is the kind of study that works great for that kind of group, isn't it?

Alberto Martini: Yeah, it is. It's mostly like a proof of concept because we wanted to see whether what I saw for a few times in practice was actually generalizable and was real. It wasn't just an idea.


Zach Klaassen: That's fantastic. So, tell us some of the key results from this study.

Alberto Martini: So the key results from the study are that lymph node invasion, we confirm what is the current rate right now in the contemporary patient population, which is approximately 13%. So, more than one patient out of 10 that is cN0, either with conventional imaging or PET PSMA, has lymph node invasion.

The main results of the study are that again, the lymph node invasion contralateral to the prostatic lobe with worse disease features, which was defined according to a definition that we actually introduced in an earlier paper that was published in the Journal of Urology actually last year, if I'm not mistaken, where we said that the dominant lobe with worse disease features is identified as the one that has the index lesion, extracapsular extension on MRI, SVI on MRI, and/or worse lesion.


In case you can't see in the paper, we had similar characteristics in the right and the left hemiprostate. We used the hierarchical criterion to select which one was the dominant lobe basically.


So for example, if we had, let's say, Gleason 4 + 4 on both sides, we defined as the dominant lobe the one that had a higher rate of positive cores, for example. So, that's what we used. Again, as I alluded earlier, we found that the metastasis contralateral to the prostatic lobe with worse disease features is extremely rare, and that's especially true in case of patients with, let's say, intermediate-risk disease according to our modified D'Amico criteria.


While it's higher of the patients that had the LNI, in case of high-risk disease, it's not unlikely to find lymph node metastases that are bilateral. So the majority of patients that have, let's say, unilateral high-risk disease, we can, in our opinion, let's say we haven't done a randomized trial, but our findings would suggest that it would be feasible to spare the lymph node dissection on that side of the prostate.


Zach Klaassen: Let's say you have a patient. You've got 4 + 4 on one side and you've got high volume 4 + 3 on the other side, probably still do bilateral on that patient?


Alberto Martini: In this case, yes, but let's say we also, once we'll discuss more about the nomograms and things like that. Let's say if we have one side that is even 3 + 4 or 4 + 3, and no cancer on the other side, the likelihood of having positive lymph nodes contralaterally to the side that has 3 + 4 or 4 + 3 is extremely rare.

Zach Klaassen: Absolutely. That's a great segue into this plot of your nomogram because I want you to walk our listeners through this, just the important points and some of the highlights because the AUC for this nomogram was 84%, which is excellent.

Alberto Martini: Yeah, thank you very much. The AUC, what is it? It's just like a measure of discrimination. What it's actually telling us here is that the variables that we selected and are mostly based on clinical knowledge are very important in predicting the presence or absence of lymph node invasion on the contralateral side. Again, here we used the PSA, the maximum diameter of the index lesion, the presence of seminal vesicle invasion. This was coded as absent or present, could have been either unilateral or bilateral.

Importantly, ECE alone on MRI, we also know that it's not very sensitive nor specific for MRI as a tool for identifying. ECE was not a strong predictor of contralateral lymph node invasion. The contralateralized group as you can see, like either having a negative biopsy or a grade group 1 or 2, basically didn't have any impact on having a contralateral lymph node invasion. Of course, this changes.


We really need to factor in the percentage of positive cores as a proxy of tumor volume on the contralateral side. Again, but in general as I said, we can say that if you have either negative biopsy of course or grade group 1 or 2, and in those cases if the biopsy positive, if the tumor is a low volume tumor, the rate of lymph node invasion is very rare contralateral to the index lesion.


Zach Klaassen: That's a great explanation. I think one thing I pulled out of your paper which I'd like you to comment too, is that if you have the lymph node invasion risk set at 1%, 27% of contralateral pelvic lymph node dissections could be omitted with only missing one lymph node invasion, correct?

Alberto Martini: That's correct. Yeah, that's correct. The lymph node invasion, actually we described how it was in the discussion. I think it was in ISoP too, which so I'm not sure, maybe it was missed on biopsy or things like that. Again yes, we could avoid lymph node dissection in one third of the patients, and then potentially spare some anesthesia, or time, and potentially reduce costs and the complications.

Zach Klaassen: Absolutely. So where do you guys take this information from here? Do you design a clinical trial? I mean obviously, surgical and clinical trials are challenging. What's the group's next steps with this data?

Alberto Martini: That's a great point. So we've tried to design a clinical trial and now we're actually debating on what would be the optimal outcome because there have been two trials in this domain. One was the one done at MSK, which was negative mostly because of a similar median count of removed lymph nodes, which was 12 versus 14 in the standard versus extended lymph dissection. Then we have the one from Brazil that actually showed that or suggested that lymph node dissection is actually helpful in case of high-risk disease in terms of PCR.


Zach Klaassen: Yes.

Alberto Martini: So we're trying to design the trial, but it's not, of course not easy.

Zach Klaassen: Right.

Alberto Martini: One because we know that PCR is not a surrogate endpoint of down the line outcomes such as metastases or survival. So, probably it's not the best endpoint. Overall survival would take forever to be evaluated. One thing that we were thinking of would be to design the trial powered on as noninferiority trial to basically show that it's not inferior in detecting lymph node invasion, which might be true.

Also, we can think of doing lymph node dissection in case if we believe that is therapeutic or just as a staging procedure. Also if we want to do it as a staging procedure, what we've shown is that even if you remove the lymph nodes on only one side, you're going to be sure that you find lymph nodes when they're positive.


Zach Klaassen: Right.

Alberto Martini: If you ask me, I believe in the therapeutic role of lymph node dissection, not in every patient, but I would do it to be able to achieve undetectable PSA after surgery. In this case, I would say that our data suggests that it's doable to achieve it if you do lymph node dissection only on one side. Again, we all know that there's no benefit in removing lymph nodes that are negative. So, I'm a true believer in what we found.

Zach Klaassen: No. It's a well-designed study and congratulations on that for sure. Just give our listeners a couple of take-home points, something they can take to the clinic tomorrow based on your study, and how to think about this as we sit and discuss this with patients.

Alberto Martini: Yeah, for sure. I would say that to discuss with patients, that the lymph node invasion contralateral to the prostatic lobe with the worse disease features is rare in a contemporary series. The presence of lymph node invasion contralateral again to the prostatic lobe with worse disease features is actually dependent on the presence or absence of tumor it create and the extent.

So I think that this is useful to counsel patients, especially those as we said that fall in the intermediate-risk category. We can tell them that we can avoid performing lymph node dissection contralaterally to a prostatic lobe with worse disease features. While if you're high-risk, in the patients that had lymph node positive in our series, most of them had bilateral lymph node invasion.

Zach Klaassen: Absolutely. That's very well said, beautiful nomogram. Congratulations on the work, and I know our listeners will look forward to this discussion. Thank you very much for your time, Dr. Martini.

Alberto Martini: Thank you very much. Of course, my pleasure.

Zach Klaassen: Thank you.