CDK4/6 and PI3K Inhibitors for HR+/HER2- Breast Cancer in the Clinic Today: Thoughts From the Experts

In this commentary, Matthew P. Goetz, MD; Sara Hurvitz, MD, FACP; and Joyce O’Shaughnessy, MD, discuss clinical data on the use of CDK4/6 inhibitors and PI3K inhibitors for patients with hormone receptor–positive (HR+)/HER2-negative (HER2-) early and advanced breast cancer from a live CCO webinar held in March 2021.

Clinical Trials: CDK4/6 Inhibitors in HR+/HER2‑ Early Breast Cancer

Joyce O’Shaughnessy, MD:
There are several clinical trials currently investigating the use of CDK4/6 inhibitors in patients with HR+/ HER2- early breast cancer (EBC).

At the 2020 San Antonio Breast Cancer Symposium, investigators presented updated data from the phase III monarchE trial (NCT03155994) that compared 2 years of standard endocrine therapy with 2 years of endocrine therapy with the addition of abemaciclib in premenopausal and postmenopausal women with high-risk, node-positive, HR+/HER2- EBC. After the study treatment, patients in both arms continued endocrine therapy for up to 5-10 years, as clinically indicated. The median follow-up was short—19 months—but for both invasive and distant disease‑free survival, there was an improvement with abemaciclib.

The phase III PALLAS trial (NCT0251339) assessed standard ongoing adjuvant endocrine therapy vs the addition of 2 years of palbociclib with standard ongoing adjuvant endocrine therapy in patients with stage II/III HR+/HER2- EBC within 12 months from diagnosis. Standard adjuvant endocrine therapy was continued for a duration of at least 5 years. The patients enrolled on this trial were mostly high risk, although less so than the enrollment criteria for monarchE. However, there was no improvement in invasive disease–free survival (IDFS) or distant disease–free survival even in those with high clinical risk (≥4 nodes involved, or 1-3 nodes with either T3 or T4 or grade 3 disease, or both) disease.

Finally, the phase III PENELOPE‑B trial (NCT01864746) enrolled patients with HR+/HER2- EBC with high risk of relapse (residual disease after ≥16 weeks of neoadjuvant chemotherapy with clinical-pathologic stage–estrogen/grade [CPS‑EG] score ≥3 or 2 with ypN+) and compared 1 year of palbociclib or 1 year of placebo in combination with standard endocrine therapy. Unfortunately, there was no improvement in IDFS or overall survival (OS) with the addition of palbociclib in this trial.

Clinical Implications

Sara Hurvitz, MD, FACP:
It was very interesting to see these data read out at the same time, but it was disappointing to see such discordant data, particularly since the data are very similar among the CDK4/6 inhibitors in the metastatic setting. There is preclinical evidence that the available CDK4/6 inhibitors are different: They have different potencies and different off-target effects. This may explain some of the differences among these trial results. Also, the length of therapy in PENELOPE-B may have affected the results. We will need to carefully review any longer-term follow-up data from monarchE to confirm the early benefit seen in this trial.

Matthew P. Goetz, MD:
I agree. When looking at the PENELOPE B data, it was interesting that there was some separation of the IDFS and OS curves early on and then those curves converged over time. This leads me to believe that the micrometastatic disease we are treating in these patients may be insufficiently controlled by palbociclib, so the survival benefit is lost once the treatment ends, or there may even be a rebound effect as has been suggested based on preclinical data.

If we compare these data with the data in the metastatic setting, I also wonder about the endocrine responsiveness of the patients enrolled in these phase III adjuvant trials. In the phase III PALOMA-3 trial (NCT01942135), palbociclib was given in combination with fulvestrant to patients with HR+/HER2- advanced breast cancer (ABC) who had disease progression after previous endocrine therapy. There were some differences in efficacy for palbociclib and fulvestrant in endocrine-sensitive and endocrine-insensitive patients. Patients who were endocrine sensitive had improved OS with the addition of palbociclib, but there was no survival advantage for patients with endocrine-insensitive disease. There may be similar findings in patients with EBC treated with palbociclib as well, but at this time, these are just theories to possibly explain the differences from these trials.

Identifying High-Risk Patients in EBC

Sara Hurvitz, MD, FACP:
When looking at these data, one important consideration is identifying patients with high-risk disease, since each of these trials had different enrollment criteria. We know that patients who have multiple lymph nodes involved are at the highest risk for recurrence, but there are patients with 1 or 2 involved lymph nodes who are also at a high risk for relapse in the long term. The 21‑gene recurrence score as well as patient age, lymphovascular invasion, lack of expression of progesterone receptor, and higher Ki-67 expression can help to identify patients with a higher risk for recurrence.

Matthew P. Goetz, MD:
The investigators in the PENELOPE‑B trial did a good job identifying a high‑risk group of patients. These patients had residual disease after neoadjuvant therapy and a CPS‑EG score ≥3 or a score of 2 with pathologic node–positive disease following neoadjuvant therapy (ypN+). Patients were already recommended for neoadjuvant therapy and had a substantial amount of disease remaining despite neoadjuvant chemotherapy. In this study’s placebo group, the 4‑year IDFS was 72.4%, indicating a very high risk. This tells us that we need to continue to study new therapies in those who have residual disease after neoadjuvant chemotherapy.

Selection and Sequence of Oral Targeted Therapy for HR+/HER2‑ ABC

Sara Hurvitz, MD, FACP:
Let’s discuss a case: An asymptomatic premenopausal woman with HR+/HER2- ABC had a bilateral mastectomy and adjuvant chemotherapy followed by paclitaxel, then tamoxifen and goserelin. After 2 years of endocrine therapy, she has recurrence with liver and bone metastases.

For this patient, I would recommend starting a CDK4/6 inhibitor. In the metastatic setting, a CDK4/6 inhibitor in combination with endocrine therapy confers a significant improvement in objective response rate compared with endocrine therapy alone and provides outcomes that are equivalent to or a little better than chemotherapy. A historical comparison of chemotherapy vs CDK4/6 inhibitor plus endocrine therapy suggest that the addition of CDK4/6 inhibitors to endocrine therapy appear to be more beneficial compared with chemotherapy.

The phase II Young‑PEARL study compared a combination of palbociclib plus exemestane and leuprolide vs capecitabine in premenopausal women with HR+/HER2- locally advanced or metastatic breast cancer who previously were treated with tamoxifen and at least 1 line of chemotherapy. Results indicated better progression-free survival with the palbociclib-containing regimen vs capecitabine, so I would be comfortable with using palbociclib for this case. Palbociclib is currently approved by the FDA in combination with an aromatase inhibitor as initial therapy for postmenopausal women with HR+/HER- ABC or in combination with fulvestrant for patients with disease progression following endocrine therapy.

For this patient case, the most data-driven approach would be the use of ribociclib with an aromatase inhibitor and ovarian function suppression. The phase III MONALEESA‑7 (NCT02278120) is the only study that exclusively looked at this approach in premenopausal and perimenopausal women with HR+/HER2- ABC. Results showed an improved OS for women who received ribociclib plus endocrine therapy and ovarian suppression vs endocrine therapy and ovarian suppression alone. Based on these data, ribociclib is approved in combination with an aromatase inhibitor for premenopausal or perimenopausal women with HR+/HER2- advanced or metastatic breast cancer as initial endocrine-base therapy. I do not recommend combining ribociclib with tamoxifen, due to the potential for QTc prolongation.

Ribociclib is also approved by the FDA in combination with an aromatase inhibitor or fulvestrant for postmenopausal patients in this setting. Abemaciclib is another FDA-approved CDK4/6 inhibitor, but the approval is in combination with an aromatase inhibitor as initial therapy for postmenopausal women with HR+/HER- ABC or in combination with fulvestrant for patients with disease progression following endocrine therapy. The abemaciclib combinations and ribociclib in combination with fulvestrant may be reasonable options for the patient case example, but they have not been investigated in a clinical trial evaluating only premenopausal and perimenopausal women in the frontline setting.

Matthew P. Goetz, MD:
I think either ribociclib or abemaciclib could be used for this patient. The best data exist for ribociclib, but there also are data on abemaciclib from the MONARCH 2 trial. In the phase III MONARCH 2 trial (NCT02107703) of women with HR+/HER2- ABC who progressed during or after neoadjuvant or adjuvant endocrine therapy, fulvestrant plus abemaciclib was compared with fulvestrant plus placebo. In this study, there was a cohort of patients who were premenopausal or perimenopausal, and a subgroup analysis showed a progression-free survival benefit for these patients. Based on these data, I would feel comfortable with either ribociclib or abemaciclib.

Joyce O’Shaughnessy, MD:
For patients with liver metastases and fast growing, aggressive disease, my choice has been abemaciclib because it had the first data to show improvement in liver metastases. Subset analyses from MONALEESA-7 and MONALEESA-3 have now also shown that ribociclib provides a survival advantage for patients with liver metastases and short disease‑free interval.

Other advantages with abemaciclib include the administration—it is given continuously without a week off—and its broader mechanism of action than the other CDK4/6 inhibitors. Abemaciclib has also shown activity in patients with either primary or secondary endocrine resistance. More recently, ribociclib has also shown good results in the primary endocrine therapy refractory population. By contrast, in PALOMA‑3, palbociclib did not confer a survival advantage in the primary endocrine therapy resistant population, although there was a survival advantage in the patients who were initially endocrine therapy sensitive and later acquired resistance.

Matthew P. Goetz, MD:
We are also doing a better job of managing the diarrhea that accompanies abemaciclib use. If you look at the monarchE data, adherence was excellent. Preemptive use of loperamide may cause constipation, so we tell our patients to take it at the first sign of diarrhea.

Joyce O’Shaughnessy, MD:
During the first 4-6 weeks of receiving abemaciclib, some women find that eating smaller, low-fiber meals that are not heavy, spicy, or fatty helps to cut down on the diarrhea. Generally, the diarrhea also improves over time.

Sequencing PI3K Inhibitors in HR+/HER2- ABC

Joyce O’Shaughnessy, MD:
Returning to the patient case, this young woman was treated with goserelin, fulvestrant, and abemaciclib, as in the MONARCH 2 study, and then she underwent an oophorectomy. She responded to therapy but, at 18 months, had progressive disease in her bone and liver and now developed right upper quadrant pain and mildly elevated liver function tests. Next-generation sequencing found a PIK3CA mutation and amplicons in FGFR1, cyclin D1, and plasma circulating tumor DNA (ctDNA) showed a new ESR1 mutation. These genetic changes are possible mechanisms of resistance while she was receiving fulvestrant and abemaciclib.

Matthew P. Goetz, MD:
Potential drivers for CDK4/6 inhibitor resistance include FGFR alterations and ESR1 and PIK3CA mutations, among others. We know that the PI3 kinase pathways are documented drivers in patients with endocrine therapy resistance. Alpelisib is a PI3Kα selective inhibitor currently approved by FDA in combination with fulvestrant for treatment of postmenopausal women and men with PIK3CA-mutated HR+/HER2- ABC whose disease has progressed on or after previous endocrine therapy. Alpelisib plus fulvestrant can also be for patients with PIK3CA-mutated HR+/HER2- ABC who have progressed on treatment with a CDK4/6 inhibitor plus endocrine therapy.

The phase II BYLieve study (NCT03056755) looked at the use of alpelisib with fulvestrant (cohort A) or letrozole (cohort B) for treatment of patients with PIK3CA mutant, HR+/HER2- ABC with progression during or after treatment with a CDK4/6 inhibitor and an aromatase inhibitor or fulvestrant, respectively. In both cohorts, the addition of alpelisib demonstrated meaningful efficacy and showed that there is a benefit to inhibiting the PI3 kinase pathway even after previous CDK4/6 inhibitor therapy.

Sara Hurvitz, MD, FACP:
For this patient, I would swap out the endocrine partner because presumably there is resistance to fulvestrant. The BYLieve trial provided promising single-arm data supporting the benefits of an aromatase inhibitor combined with alpelisib in patients who had received previous CDK4/6 inhibitor and fulvestrant.

Even with the right upper quadrant pain and elevated liver function tests, I would still recommend alpelisib if the patient’s symptoms are not worsening. But, because of the toxicity profile of alpelisib, we need to check the patient’s fasting plasma glucose and start patients on antihistamines before starting alpelisib. Then, we frequently monitor fasting plasma glucose levels and check in regarding diarrhea and rash. We are learning to better manage the glucose issues with alpelisib, and SGLT2 inhibitors are being used to help patients secrete the extra sugar and prevent elevated insulin levels. If a patient has uncontrolled diabetes or is at high risk for developing some of the adverse events associated with alpelisib, then in some cases, I would try capecitabine first.

Joyce O’Shaughnessy, MD:
I agree. The phase III SOLAR-1 trial evaluated the efficacy of alpelisib plus fulvestrant compared with placebo plus fulvestrant in HR+/HER2- ABC after progression on a previous aromatase inhibitor and included a cohort of patients with PIK3CA mutations and a cohort without PIK3CA mutations. In the cohort of patients with a PIK3CA mutation, addition of alpelisib resulted in prolonged median progression-free survival when compared with the control. In a subgroup analysis of this cohort, patients did well regardless of lung or liver metastases, bone only disease, or previous CDK4/6 inhibitor therapy.

In an exploratory analysis of SOLAR-1, outcomes in patients with and without PIK3CA alterations as retrospectively detected by next-generation sequencing in ctDNA and breast tissue was assessed. This analysis revealed a discordance between detectable PIK3CA alterations in ctDNA vs tumor tissue and showed that 38% of patients who did not have a PIK3CA mutation in their plasma ctDNA had a PIK3CA mutation in their tumor tissue and would benefit from alpelisib therapy. Therefore, it is important to remember that if you do not find a PIK3CA mutation by ctDNA analysis, you should still check the primary breast tissue for mutations.

I also want to emphasize that the A1C has to be <6.5% for patients receiving alpelisib. Otherwise, it will be difficult to manage blood sugar levels. If the glucose level is >160 mg/dL, I prescribe metformin. So far, I have not had to prescribe SGLT2 inhibitors.

Using CDK4/6 Inhibitors in Patients With Indolent Disease

Joyce O’Shaughnessy, MD:
For our second case, in 1998, a 48‑year‑old, premenopausal woman was diagnosed with a right stage I estrogen receptor–positive (ER+) breast cancer. She had 6 rounds of cyclophosphamide/methotrexate/5-fluorouracil and radiation, and she subsequently declined endocrine therapy. In 2011, at the age of 61 years, she returned with a chest wall recurrence involving the pectoralis muscle, and biopsy shows ER+/progesterone receptor–positive/HER2- disease and a low Ki-67 expression (5%). She was treated with exemestane for 1 year with a very good clinical response and then underwent a wide margin surgical excision but had 5 cm of disease with extensive positive margins in the muscle. A postsurgery bone scan revealed bone metastasis in the sternum. She had not had a baseline bone scan before starting the exemestane, and then a biopsy of the bone shows ER+/HER2‑ disease. The patient chose to continue therapy with exemestane and initiate a bone agent. She had frequent scans showing sclerosis/healing of the sternum, and her chest wall scar had stable nodularity with no evidence of disease elsewhere. When palbociclib was approved in 2015, the patient declined to start a CDK4/6 inhibitor and did well until 2020 when she had clear progression involving the chest wall scar. A biopsy showed the disease was still ER+/HER2-, but it was now progesterone receptor negative. There was no progression elsewhere, just in her scar.

Sara Hurvitz, MD, FACP:
I am a huge proponent for starting a CDK4/6 inhibitor in almost all cases in patients with metastatic ER+ breast cancer. In this case, the patient did well before any CDK4/6 inhibitors were approved by the FDA. The problem is that we do not know how to pick the patients who are going to have a prolonged, durable response or stable disease without adding additional therapy. The FDA approval of palbociclib came when the patient had already been receiving exemestane for several years, and given her lifestyle and travel, she did not want to add any additional therapy that would require more medical visits or blood tests.

It took a long time to convince her to add a CDK4/6 inhibitor and switch over to fulvestrant. Ultimately, we did and the response lasted approximately 1 year. Patients who have enjoyed many years with single‑agent endocrine therapy may not be willing to give up the tolerability associated with that therapy. This case highlights the fact that there are patients who have a more indolent disease course and can do well with single-agent endocrine therapy, but we are not able to select who these patients are at the beginning of therapy.

Matthew P. Goetz, MD:
This is an excellent point. I will point out that the presence of an ESR1 mutation suggests resistance to single-agent endocrine. However, some data suggest that a CDK4/6 inhibitor combined with endocrine therapy may be just as effective as chemotherapy in patients with ESR1 mutations. There are no data directly comparing CDK4/6 inhibitors to advocate one over another, but in the presence of an ESR1 mutation, a patient may not do as well with single‑agent endocrine therapy. 

Your Thoughts
What questions do you have about selecting CDK4/6 inhibitors for your patients with HR+/HER2- breast cancer? Please answer the poll and post your thoughts and questions in the discussion box below.