Choosing Biomarker Assays After Metastatic Relapse of ER+, HER- Breast Cancer

Mark Pegram, MD

Suzy Yuan-Huey Hung Endowed Professor of Medical Oncology

Stanford University School of Medicine

Associate Director of Clinical Research

Stanford Comprehensive Cancer Institute

Stanford, CA

CASE SUMMARY

A 56-year-old postmenopausal woman presented with a palpable right breast mass with no clinically abnormal axillary lymph nodes. A core biopsy led to diagnosis of grade 2, estrogen receptor–positive (ER+)/progesterone receptor–positive (PR+) invasive ductal carcinoma (IDC) with a HER2 immunohistochemistry (IHC) score of 0 and Ki-67 score of 33%. Lumpectomy and sentinel lymph node (SNL) biopsy results showed tumor measuring 3 cm, grade 2 IDC, and 2 SNLs negative for malignant cells. Her 21-gene recurrence score was 27.

The patient received docetaxel (Taxotere) and cyclophosphamide (Cytoxan) for 4 cycles followed by radiation therapy, and she completed 5 years of adjuvant aromatase inhibitor (AI) therapy. Three years later, she reported right-sided abdominal pain and mild nausea. A CT scan of the chest, abdomen, and pelvis showed 3 suspicious liver lesions (right lobe; largest measuring 2 cm). Liver biopsy results showed ER+/PR+ IDC with a HER2 IHC score of 0. Comprehensive molecular testing results from tissue biopsy showed no actionable alterations. AI therapy plus palbociclib (Ibrance) was initiated.

The therapy was well tolerated by the patient, with low-grade neutropenia that did not require dose modification. Twenty months later, follow-up imaging shows enlargement of liver nodules and 2 new lung nodules (largest measuring 0.9 cm). Her ECOG performance status is 0.

MARK PEGRAM, MD: Comprehensive profiling via Guardant 360 CDx was the most common answer, followed by tissue biopsy. One person said comprehensive profiling using a liquid biopsy from another assay. I think that’s the right answer. You need to know, among other things, what the PI3 kinase mutational status is because those patients might be treated with fulvestrant [Faslodex] plus alpelisib [Piqray], for example. You might have some enlightening feature of the tumor that [you wouldn’t have known about] if they haven’t had DNA sequence analysis previously on their tissue and only standard immunohistochemistry assays. I think I’m going to do the same thing.

At our center, we most frequently use Guardant 360, but I think all the commercial diagnostic houses now have ctDNA [circulating tumor DNA] assays available and they have similar geneless, as does Guardant. So I don’t think it matters much which vendor you use as long as you remember to get them. They’re all going to [assess for] ER mutations, HER2 kinase domain mutations, PI3 kinase [PI3K] mutations, ATK mutations, etc.

DISCUSSION QUESTION

What would be your next step for this patient?

Do you routinely re-biopsy/order genomic testing at progression on endocrine therapy plus CDK4/6 inhibitor?

PEGRAM: What platforms or methodologies are you using at your center?

KELLY E. MCCANN, MD: I use a lot of Tempus [xF] testing because the financial aspect of it is more open. I’ve been burned on some of the other ones, but they’re all good.

PEGRAM: Tempus certainly has a very fine product [with] high quality control. I’ve been to their shop a few times in Chicago. I had meetings with the chief executive officer and they have a very robust shop there, so nothing wrong with that one. Anybody else using anything else? Anyone using FoundationOne, for example? They also have a ctDNA platform. What about [the Signatera assay from] Natera?

SWARNA CHANDURI, MD: We started using Natera. We used to use Guardant 360, but recently we started using the Natera.

PEGRAM: Natera is neat because they sequence 16 of the most frequently mutated genes in the patient’s tumor from their diagnostic biopsy. And then since they know exactly what they’re looking for, they make custom primers for just those mutations that are unique to each patient. By using these unique primer pairs, they can do incredible depth of sequence penetration, something like 100,000-fold, so it’s extremely sensitive.

Since they only use the mutations that are in the patient’s original tumor, if the patient acquires, let’s say, a PI3 kinase mutation later on, you would miss that because they only look for the original gene set with the Natera assay. They don’t look at a set number of genes that are commonly altered in everyone’s tumors necessarily. And sometimes PI3 kinase mutations can be acquired [but be missed] initially in the time of metastatic relapse. So that’s an interesting twist you might need to consider if you don’t find a PI3 kinase alteration. You might want to re-sequence using a different platform with the Natera assay in that case.

Do you routinely rebiopsy any of these cases after endocrine therapy and CDK4/6 inhibitor or are you happy with the liquid biopsy?

ARATI CHAND, MD: I usually don’t biopsy now with the Guardant 360 being the companion diagnostic testing for the ESR1 mutation.¹ It’s been easy to get that approved post progression of CDK4/6 inhibitor and AI especially.

I recently had a patient whom I checked for PI3 kinase on the primary [tumor]. It was negative, but then I sent out the Guardant 360. She had ESR1 and PI3 kinase. How often do you see these acquired mutations come in…circulating tumor cells even if the primary doesn’t have it? Also, how do you sequence; do you [target] ESR1? Would you use elacestrant [Orserdu] first or would you use alpelisib first? How do you sequence these agents?

PEGRAM: Since both ER mutations and PI3 kinase mutations are common in metastatic disease and also more common than in the primary, certainly with ER mutations, the primary tumors untreated have less than a 1% frequency of ER mutation. But, in metastatic disease, if they’ve had a prior AI, you can find it in perhaps a third and something like 40% of the patients will have PI3 kinase mutations with metastatic breast cancer. The coincidence of the 2 is going to happen.

You had a case that had both ER mutation and PI3 kinase mutations, both of which now are actionable. I tend to use therapeutic index in metastatic disease for clinical decision making, that is efficacy divided by toxicity. Using that strategy, I’ve had challenges giving alpelisib. It’s a tough drug to give. I had 2 patients with glucoses over 600 mg/dL who weren’t diabetic, for example. Personally, I would probably treat the steroid receptor mutation first and reserve the alpelisib for later. But there’s no data on this whatsoever…so you can present both to the patient to see if they have a preference one way or the other. Perhaps they’ll be more attracted to the concept of targeting an oncogene rather than a steroid receptor. A balanced discussion is perfectly reasonable, but probably if push comes to shove and they ask me what I would like to do, I’d probably use the less toxic of the 2 first in general. That’s my personal preference, but it doesn’t make it right or wrong.

KAMALESH SANKHALA, MD: On someone who had no biopsiable tumor subsequently and if we are doing Natera to look for acquired mutation, do you recommend that we should do Guardant 360 or some other ctDNA?

PEGRAM: You can send a metastatic biopsy to Natera and they’ll re-sequence it and then you’ll find any new alterations that have emerged. But they need a biopsy specimen, I believe, to make that determination. Whereas, Guardant 360 is a fixed gene list and they’ll sequence everything on that list but they won’t sequence anything else. Both of them can be complementary, likely, but to catch the actual mutations that can be acquired since diagnosis, only Guardant would capture those.

_^Reference:

_{1. Guardant Health receives FDA approval for Guardant360® CDx as companion diagnostic for Menarini Group’s ORSERDU™ for treatment of patients with ESR1 mutations in ER+, HER2- advanced or metastatic breast cancer. News Release. Guardant Health, Inc. January 30, 2023. Accessed June 7, 2023. https://bit.ly/3HGF5hu}