Sara Tolaney, MD, MPH:
At this early timepoint, NATALEE shows that 3 years of adjuvant ribociclib plus an NSAI significantly improved iDFS and, of importance, also improved DDFS vs NSAI alone in patients with intermediate-risk to very high–risk HR-positive EBC.¹ One caution is that elevation of liver enzymes was a major reason patients discontinued ribociclib, and physicians will need to monitor that carefully. We also need to monitor patients for neutropenia and QTc issues related to ribociclib, although those did not emerge as notable problems in this trial.

Even though these data are early, it is useful to consider whether we would want to start prescribing ribociclib if the FDA grants approval in this setting. I have mixed feelings. For one, I am very impressed with the benefit that we see to date, a 25% relative risk reduction in this population that includes both very high–risk and intermediate-risk disease. However, I would like to see the data broken down more granularly so we can understand the absolute benefit in different subpopulations, such as those with stage IIa disease who already have a lower absolute risk of recurrence.

We will also watch to see how these data mature as more patients complete their 3 years of ribociclib. In the monarchE trial, the iDFS curves have continued to separate, showing greater benefit with abemaciclib over time, and we could reasonably anticipate a similar phenomenon with ribociclib.^2,3 But we will need to see that confirmed. Furthermore, given the notable rate of discontinuations in NATALEE—many related to elevated liver enzymes—we will also need to weigh the risks vs benefits for patients.

At this time, for patients with very high–risk disease who meet monarchE eligibility criteria, I would choose to treat with adjuvant abemaciclib. We have more mature data with abemaciclib; the longer follow-up with patients who have completed 2 years of treatment is showing continued improvement in the hazard ratio for iDFS with time, suggesting there might even be some carryover effect after completing abemaciclib.^2,3 The 4-year follow-up data show an approximately one third risk reduction in iDFS events, demonstrating that abemaciclib is clearly a beneficial adjuvant therapy for those with high-risk disease. Based on monarchE, the abemaciclib label has also been updated to drop the Ki-67 threshold requirement.⁴

If the benefit of adjuvant ribociclib is confirmed in subsequent analyses and studies, these results could broaden the population eligible for CDK4/6 inhibition to include patients with higher-risk, node-negative disease who are ineligible for abemaciclib. I would consider ribociclib for those with intermediate-risk, T2N0 disease who have very high genomic assay scores and high clinicopathologic risk features, but do not meet monarchE eligibility—for example, a patient with a 4.5-cm, high-grade tumor with a genomic assay score of 35.

With longer follow-up of NATALEE, I hope to see ribociclib as an alternative adjuvant option for patients to reduce the risk of recurrence.

Erika Hamilton, MD:
It is important to consider the broader population enrolled in the NATALEE trial compared with monarchE, including some with intermediate-risk as well as high-risk disease. Recall that 27% to 29% of the patients on NATALEE did not have lymph node involvement.

However, as Dr Tolaney mentioned, the investigators did not provide a breakdown of benefit according to risk. A patient with 6 positive nodes might be expected to have greater benefit than someone with no positive nodes, but we need to see the data to be able to predict an individual patient’s benefit based on their specific risk factors. With 80% of patients still receiving ongoing ribociclib, this dataset is very immature. I hope we will see more data broken down by risk group in the future.

The use of a 400-mg dose of ribociclib in NATALEE, compared with the 600-mg dose we use in the metastatic setting, was an interesting twist. The rates of dose dependent toxicities were lower with the 400-mg dose of ribociclib when we compare across trials. However, neutropenia was still quite common, with more than 60% of patients experiencing any grade neutropenia and more than 40% with grade ≥3 events.

Of course, ribociclib is not currently approved by the FDA in this setting, and we will watch to see how the FDA approaches these data. If ribociclib is approved, it will be interesting to see if the indication is for the whole population included in the NATALEE trial or just for a subset of patients, as we saw with abemaciclib initially. Again, we will need more data related to individual risk groups to know how to talk with our patients.

Overall, I find these results very encouraging. I think this may give more patients access to a CDK4/6 inhibitor in the adjuvant setting, which is important. These are patients who might relapse and live the rest of their lives with metastatic disease; by treating them with CDK4/6 inhibition while they are still in the adjuvant setting, we might be able to convert some of them to cure. This could have a big impact.

Erika Hamilton, MD:
My first takeaway from this analysis is that our older patients appear to derive benefit from abemaciclib comparable to that of our younger patients. Sometimes, we may think that older patients are more likely to have more indolent cancers and may not need aggressive therapy to achieve a good outcome. The results of monarchE tell me that within a patient population selected based on clinical high risk characteristics—positive nodes, tumor size, tumor grade—older patients have an equally poor outcome and can benefit from drugs like abemaciclib.

My second conclusion is that AEs were not very different between younger vs older patients. For example, neutropenia, liver enzyme increases, venous thromboembolism, and ILD were all numerically similar between the age subgroups. The older subgroup had 5% less any grade diarrhea, but 5% more grade 3 severe diarrhea. Grade 3 fatigue occurred in 2% of patients aged younger than 65 years and 6% in those aged 65 years or older.

Despite these patient subgroups deriving similar benefit from abemaciclib and seeming to tolerate the drug similarly, older patients had more dose reductions and more than double the rate of abemaciclib discontinuations before their 2 years of therapy were completed. Surprisingly, more than one half of those who discontinued abemaciclib, regardless of age, did so without ever trying a dose reduction.

For this reason, we looked at efficacy by relative dose intensity. We defined 3 groups: patients who received at least 93% of the planned abemaciclib dose, a middle group that received 66% to 93% of the intended dose, and a low group that received less than 66% of the planned dose because of dose reductions, skipped days, or other reasons. Looking at 4-year iDFS rates across these groups, we did not see any difference in abemaciclib efficacy by relative dose intensity.

Taken together, these results tell me that physicians should emphasize to patients upfront that dose reduction is an option if they have trouble tolerating abemaciclib, and a lower dose does not seem to lead to worse outcomes. Sometimes, if a patient experiences an AE, they may just stop taking the drug without letting their physician know. We may not find out until the patient comes back into the clinic.

Historically, I think both physicians and patients have been apprehensive about trying dose reductions. If a drug helps you, it is easy to think that taking less of the drug might help you less. But it is important to remember that, unlike chemotherapy drugs that are often dosed based on weight or body surface area, many oral drugs are given at the same starting dose regardless of body size. Yet, the reality is that the same dose is probably not right for every patient.

I think it will help if we can set the expectation upfront that some people may require lower doses if they have AEs. Seeing in a study like this that patients can reduce their dose and still have equally good outcomes gives us the reassurance that this strategy is appropriate to use in the clinic. I would like to see fewer people discontinue abemaciclib without trying a dose reduction first.

Sara Tolaney, MD, MPH:
Remember that approximately 18% of patients overall and 38% of patients aged 65 years or older discontinued abemaciclib before completing 2 years of treatment. The analysis looking at factors associated with relative risk in time to discontinuation of abemaciclib suggests that older patients and those with comorbidities are more likely to discontinue treatment. But again, this analysis was done when people did not fully realize the importance of dose modification. It serves as another reminder that dose modifications can make abemaciclib tolerable and useful, even for older patients with more comorbidities.

Sara Tolaney, MD, MPH:
This study reported good efficacy and safety outcomes with PH, with an outstanding 3-year iDFS rate among those patients who had a PET response and achieved pCR.⁶ The challenge is whether this approach is ready for broader use. At this point, I feel we would not want to use this approach outside of a clinical trial because we still are only seeing a pCR rate of approximately 40%. That means that 60% of patients are escalating to full TCHP. I think this is a very important strategy, but we need to tailor it further so that we can achieve a higher pCR rate with imaging response. Alternatively, we could try using biomarker-selected therapy—for example, using the HER2DX genomic test.

Erika Hamilton, MD:
I did not find these results to be practice-changing; rather, they were practice-affirming because these data support our current understanding of risk and age. What these results tell us is that when we have a very young woman in her 20s or 30s, we should realize that this patient probably has a higher-than-average risk. Many healthcare professionals already knew that, but it is helpful to see confirmation in this analysis.

Erika Hamilton, MD:
Continuing with the theme of ovarian suppression, Gray and colleagues¹² presented a meta-analysis of nearly 15,000 premenopausal patients with EBC across 23 trials to examine the effects of ovarian ablation or suppression on breast cancer recurrence and mortality. The analysis looked at outcomes by age and nodal status and considered whether patients had ablation vs suppression with a gonadotropin hormone-releasing hormone agonist such as goserelin or leuprolide.

Erika Hamilton, MD:
That difference in outcomes by method gives me pause because an active area of discussion right now is how thoroughly pharmacologic approaches are suppressing ovarian function. For example, leuprolide can be given via monthly shots or a 3-month depot, and we have seen data suggesting that some patients break through that 3 month shot.¹³ For me, this result is a bit thought-provoking, raising the question of whether the shots are as effective as ablation at blocking ovarian function.

In clinic, we typically present both options to patients. Some patients prefer shots to avoid surgery. Others find laparoscopic surgery more appealing than monthly shots for 5 years. Patients who want to preserve fertility must opt for pharmacologic suppression rather than ablation, of course. When selecting a gonadotropin hormone-releasing hormone agonist, I have not seen much difference between goserelin and leuprolide. I think it probably does not matter which agent you use, and the options may be dictated by insurance, a trial protocol, or the country.

Typically, I start treatment with the shots for several months to see how well the patient tolerates ovarian suppression. You can always stop shots if somebody is not tolerating it well, but you cannot undo a surgical removal. After a few months, many of my patients elect to undergo surgery so they can be done with the shots.

Unfortunately, there are also insurance and cost considerations. Often, patients incur co-pays each time they come into the clinic for a shot. A one-time surgery can sometimes be financially advantageous compared with years of monthly shots. The shots given every 3 months can reduce visits somewhat, but the possibility of breakthrough ovarian function makes me less comfortable with that option.

Erika Hamilton, MD:
As our understanding evolves about how different patients respond to various treatment options for EBC, it is interesting to consider how HCPs select treatments for individual case scenarios. Clinical Care Options developed an online decision support tool that provides case-specific guidance for (neo)adjuvant treatment of HER2-negative EBC based on recommendations from a panel of 5 breast cancer experts. In this study, the investigators looked at intended treatment patterns among HCPs who used this decision support tool and how they compare with the expert recommendations for particular case scenarios involving high-risk patients.¹⁴

Erika Hamilton, MD:
The comparison focused on 3 case scenarios. Case 1 was a patient with HR-positive, HER2-negative disease and no known deleterious BRCA mutation.¹⁴ This is a patient who would fit the high-risk criteria for the monarchE trial we discussed earlier. Case 2 was a patient with BRCA-mutated triple-negative breast cancer (TNBC) who had residual disease after neoadjuvant chemotherapy and/or pembrolizumab. Case 3 was a patient with TNBC who had no residual disease after neoadjuvant treatment.

The expert consensus rate was very high for these cases. All experts recommended abemaciclib plus ET for case 1. For case 2, 4 of 5 experts recommended adjuvant olaparib plus pembrolizumab for case 2. All experts recommended observation only for case 3.

There was a notable degree of discordance (40%-63%) between experts and the community HCPs, with much more variation among intended treatment plans reported by community HCPs for each case. For example, for the patient in case 1 with high-risk, HR-positive EBC, just 37% of surveyed HCPs indicated they would use the expert-recommended treatment of abemaciclib with ET. Nearly half indicated they would instead use chemotherapy and ET.

For the patient in case 3 with TNBC and no residual disease, only 40% of community HCPs chose the same treatment plan as the experts, which was observation without any adjuvant therapy. The other 60% of community physicians either planned to treat with pembrolizumab or capecitabine or indicated they were uncertain how to treat.

Erika Hamilton, MD:
There may be several reasons for this discordance between the expert recommendations and how providers in the community are treating patients with EBC.

First, it is important to remember that many community oncologists see patients with not just breast cancer, but also pancreatic, lung, colorectal, prostate, and other cancers. It can be hard to keep up with the literature for a single type of cancer, never mind so many different malignancies, even for those of us who do specialize.

Second, the breast cancer field moves quickly. Data from multiple studies often come out more or less simultaneously. It is not inherently obvious what to do with results when 2 drugs have been recently approved and because the studies were ongoing concurrently, neither study addresses the use of the other drug. We can see this in case 2; this is a situation where adjuvant PARP inhibitors were approved for patients with residual disease after neoadjuvant therapy. At the same time, data were emerging with pembrolizumab. A treating oncologist would face several difficult questions. Should you give a PARP inhibitor with pembrolizumab? Based on results from the CREATE X trial with capecitabine, would you give capecitabine with pembrolizumab?¹⁵ Or would you switch to a PARP inhibitor or capecitabine and stop the pembrolizumab completely? These results were coming out at approximately the same time, and no trial has been done to prospectively answer any of these very nuanced questions. In this case, not even all the experts agreed: 4 out of 5 recommended treating with olaparib plus pembrolizumab, and the other recommended olaparib alone.

The variety of responses we see from the community HCPs across all the cases included numerous people indicating they are uncertain how to treat—including nearly a quarter of the surveyed HCPs on case 3. I think we should normalize that feeling of being uncertain. If I had to keep up with all of the literature coming out on so many different tumor types, I would be uncertain a lot, too. Given how quickly treatment options evolve, we are unlikely to have trial data available to guide how to combine different regimens.

CME programs can discuss these clinical dilemmas, but it is tough to integrate new data into treatment plans. When a provider is uncertain of what to do in a particular scenario, I recommend consulting the newest National Comprehensive Cancer Network guidelines and reaching out to experts in the specialty for second opinions.