Joyce O’ Shaughnessy, MD: Could this approach help us with the patient we just saw? Is there a role for short-term endocrine therapy to help us figure out the biology and need for chemotherapy in these gray zone cases? What about longer endocrine therapy? Where might we utilize that? I’d love to hear from Sunil. Can we do that? How reliable is a suppression of Ki67 of less than 10%? If we give longer duration endocrine therapy, it takes a long time for us to know if it’s working. Should we get an intermediate Ki67? Should the cutoff be less than 10%? In the ALTERNATE trial, complete cell cycle arrest is 2.7%, but I don’t know if you can read that amount out. I’d love to hear Sunil’s thoughts. But let me ask the clinicians first. Should we be doing this? Should we be getting some preoperative endocrine therapy? Heather, what do you think?
Heather McArthur, MD: I love this idea. It’s very much aligned with where the field is moving in general, which is looking at response and biomarkers to predict who needs more and which patients can have a more de-escalated or optimized approach. I love this idea in general. It’s worth noting that there were no patients who converted from low to high. If you had a low Ki67 at baseline, there would be a limited role to re-evaluating, and maybe those patients wouldn’t benefit from this type of strategy. Then it’s limited to those who have high scores at baseline. It’s provocative. I’m not sure I’m ready to use this strategy in the clinic tomorrow, especially given all the issues around Ki67 testing that we’ve talked about already.
You mentioned the ALTERNATE trial. That was anastrozole [Arimidex] vs fulvestrant [Faslodex] vs the combination in the neoadjuvant setting for 24 weeks, with less than 1% pCR [pathologic complete response] rate for all of those groups. I don’t know what the right hormone therapy is to apply in this setting, or what the optimal duration would be, so I wouldn’t be prepared to do this in the clinic off study. But I’d love to see more of these tailored study designs and would be happy to participate in those.
Joyce O’ Shaughnessy, MD: Thanks. How about you, Massimo? Do you guys do this at all? Do you think we should be doing some of this?
Massimo Cristofanilli, MD, FACP: Most of the time, we use neoadjuvant endocrine therapy in all the patients when we want to reduce the tumors but we don’t want to use chemotherapy because of the toxicity. In general, this is the use in the clinic in practice in the United States. This patient didn’t have a large tumor. She had a small tumor. She’s relatively young. In the context of a study, 2 or 3 weeks would be appropriate. It would be difficult to make the patient wait longer in general clinical practice because these patients obviously want to pursue primary treatment as soon as possible.
Part of the issue is also the heterogeneity in the Ki67. You can do a biopsy. You need to be careful that you biopsy the same area of the tumor so that you can trust that if you have the second measurement less than 10%, you’re properly selecting patients who continue endocrine therapy. It may be highly variable. This is 1 of the major issues when you do biomarkers: the accuracy, the producibility, etc. The answer from the German study looks perfect. I don’t think I’m ready to do this in everyday practice. I’ll certainly welcome a study and participate. We can certainly do that with specific questions and maybe together with a Ki67 try to find another way to measure biological response to treatment.
Joyce O’ Shaughnessy, MD: Thanks, Massimo. Sunil, what do you think about Ki67 in this setting?
Sunil Badve, MD, FRCPath: First, it’s the best possible marker that we have. All other cell cycle markers that we have looked at haven’t worked well in predicting response to endocrine therapy or CDK4/6 inhibitor. At last we have something, as flawed as it might be. But it’s a good marker that one needs to tame and use in a consistent manner. That’s my first comment. The second is now we’re in the COVID-19 era where a lot of patients may not be willing to come inside our hospital for surgery. I’ve seen people use short-term endocrine therapy to tide the process until they can come in and feel comfortable enough to get surgery and chemotherapy, as the case might be. That’s an area that one can definitely use these short-term therapies on.
Lastly, Ki67 is like any marker. As Heather mentioned, this German study and the POETIC study show that none of the laws went into high on excision, clearly documenting consistency in the analysis that was done between the biopsy and the excision. We may blow up as much as we want about the marker not being consistent or have concerns about it, but only after we do these kinds of analyses will we know what the problems are. At least in the German hands and the British hands, the markers appear fairly consistent in spite of all the concerns that we have raised.
Joyce O’ Shaughnessy, MD: Thank you so much. We have an innovative surgeon where I work, and I’m OK with this. If I got the upfront biopsy and the patient hasn’t had any endocrine therapy, that’s a good sample for 21-gene or 70-gene signature. It’s unadulterated. I can get good information. If it comes back and I see the patient and I’m scratching my head and don’t know if this is the biology that’s going to benefit from chemotherapy, they’re going to go to surgery first, we’re going to get all the information, and we’re going to meet again and put our heads together. But so far, the biology is looking somewhat indolent.
My surgeon has started putting the patient on endocrine therapy for a few weeks before surgery, and then come back and we have that Ki67 on endocrine therapy. I’m happy to have it. It’s 1 more piece of information. As long as it doesn’t get in the way of getting a prognostic signature or interfere in any way with chemotherapy, which I doubt would be the case because usually it’s in the context of a more indolent type biology. And particularly for that premenopausal patient, where she may have 1 node that looks suspicious on ultrasound and she may be in that 1 to 3 node range where the biology looks very indolent, you don’t want to give her chemotherapy. It’s 1 more piece of information. We’re talking to Nadia Harbeck, MD, PhD [University of Munich]. They find this very useful over in Germany. It’s interesting. I don’t see a downside to doing it.
At any rate, thank you guys very much. We’ve hit on some good controversies here and helped us a little. These things can help our tumor boards all the time. Sunil, I want to thank you. This whole Ki67 discussion and the vagaries of it is a steep learning curve for a lot of us. It’s good to hear you think that it’s been around a while, and once there’s some consistently applied methodology, the utility will be there for us. Thank you all very much. It’s been a pleasure talking with everybody. I hope this has been very helpful to everybody for your practice.
Transcript Edited for Clarity