Barroilhet Reviews New Endometrial Cancer Data of Later-Line Therapy

Lisa Barroilhet, MD, MS

Clinical Associate Professor

Director of Clinical Research

University of Wisconsin School of Medicine and Public Health

Madison, WI

CASE SUMMARY

August 2021

• A 64-year-old postmenopausal woman presented with abnormal uterine bleeding lasting 4 months.
• She noted that she underwent menopause aged 55 years.
• She is a widow, has no children, and lives alone.
• Medical history: arthritis, obesity (body mass index of 40), hypertension well controlled with medication
• ECOG performance status: 1
• Endometrial biopsy: stage IA, grade 1 endometrial cancer
• The patient was counseled on surgical options then scheduled for total laparoscopic hysterectomy with bilateral salpingo oophorectomy and sentinel lymph node dissection.
• Immunohistochemistry/molecular testing results:
  • Mismatch repair deficient (dMMR)
  • Microsatellite instability-high (MSI-H)
  • Estrogen receptor positive (score of 3+)
• The patient was counseled on surveillance every 6 months for a year and then annually thereafter.

August 2022

• She reported intermittent pelvic pain over prior 4 weeks.
• Chest, abdomen, and pelvis CT scan suggested relapsed/metastatic disease with involvement of 1 right external iliac lymph node.
• Carboplatin/paclitaxel (6 times every 4 weeks) was administered, chemotherapy was well tolerated, and a complete response (CR) was recorded at end of regimen.

April 2023

• Disease relapse was documented on routine follow-up.
• Chest, abdomen, and pelvis CT scan showed heterogeneously enhancing mass in right suprarenal space, multiple bilateral pulmonary nodules, and a new right internal iliac lymph node (in addition to the previously observed cancer-positive lymph node).
• Fine-needle aspiration of the suprarenal mass confirmed metastatic endometrioid adenocarcinoma.
• The patient was counseled about systemic therapy options, during which she repeatedly expressed concerns about adverse events (AEs).

Targeted Oncology: What immune checkpoint inhibitor (ICI) data should be taken into consideration for this patient with endometrial cancer and dMMR, MSI-H disease?

BARROILHET: NRG-GY018 [NCT03914612] created a lot of attention when it was published in early 2023.¹ The NRG cooperative group designed the study looking at carboplatin and paclitaxel [Taxol] vs carboplatin, paclitaxel, and pembrolizumab. These were patients with advanced endometrial cancer, stage III or IV or recurrent disease, so the population typically getting chemotherapy. Patients had measurable disease if they had stage III or IVa disease; if they were stage IVb or recurrent, they did not have to have measurable disease to be included. Investigators did include both MMR proficient [pMMR] and dMMR patients in the study. They were stratified based on that result. The majority of patients were pMMR and about a third were dMMR.

The standard arm was placebo controlled, and the other arm included pembrolizumab [Keytruda], and then the patients would go on after completing their 6 cycles of carboplatin and paclitaxel to receive maintenance pembrolizumab for a total of 12 months. They were stratified based on their MMR status, their performance status, and the presence of measurable disease. They did allow performance status of 2, which I thought was great.

What were some of the baseline characteristics of this population?

The median age is what you’d expect; endometrial cancer in our country is typically diagnosed in women in their 50s, 60s, and 70s, which is reflected here. It was a largely White population. The majority of patients had performance status of 0 or 1, but they did capture some people with worse performance status. The majority of patients had endometrioid subtype, with a number of grades represented. There were serous cancers, not a lot. There were a couple of differentiated and clear cell cancers also. Some patients had had prior chemotherapy since some of the patients included in the study were recurrent, not just treatment naive. Radiation was allowed, and there were quite a few patients in the study who had received radiation or were even getting concurrent radiation. Most patients with uterine cancer had had surgery at some point.

How did patients do on the NRG-GY018 study?

Looking at the progression-free survival [PFS] in both cohorts [dMMR and pMMR], this was a preplanned analysis; it wasn’t a post hoc exploratory analysis. The study size was calculated so these cohorts could be analyzed separately. In the dMMR cohort, there was a very robust difference in PFS. At the time of the publication, the median PFS in the immunotherapy group had not yet been reached compared with 7.6 months in the placebo group who received paclitaxel/carboplatin [HR, 0.30; 95% CI, 0.19-0.48]. In the proficient cohort, this was surprising to many. There was a pretty robust difference as well: 13-month median PFS with the addition of pembrolizumab compared with 8.7 months with paclitaxel and carboplatin alone [HR, 0.54; 95% CI, 0.41-0.71]. This is one of the studies I remember where I was when I was reading it. I thought things are about to change a lot.

In terms of safety, what were the notable AEs on this trial?

I think AEs are important to consider, especially when we’re talking about integrating an additional medication into up-front therapy. These are often patients for whom quality of life is very important. This is their first systemic therapy. It’s what I think we have all seen in our practices: fatigue, neuropathy—especially anytime we give paclitaxel—anemia, nausea, and then bowel changes. The grade 3 or more events were uncommon; anemia was the most common grade 3 or greater AE reported. Then there were blood count abnormalities, weight loss, and rash.

With the AEs of special interest, all of us reading studies that include ICIs or other forms of immunotherapy are hyper-focused on these autoimmune conditions that can be caused by drugs in this class. Infusion reactions were not super common. But hypothyroidism and hyperthyroidism were both seen. There were no grade 3 events reported but certainly patients were started on [thyroid] supplementation.

Are there other new ICI options for this patient population?

There was a phase 3 study that included dostarlimab [Jemperli], the RUBY trial [NCT03981796], which was presented this year.² It had overlapping eligibility to NRG-GY018. We’re talking about recurrent or primary advanced endometrial cancer; they did include stage II patients in the study since some of those patients do get chemotherapy.

[They were randomly assigned] 1:1, similarly they were stratified not by MMR, but instead [by] MSI status, which is essentially a surrogate. They were stratified based on prior radiotherapy and also on disease status, so burden of disease. The 2 arms were dostarlimab, which was given at 500 mg—it is an [intravenous] medication—at area under the curve of 5, and then our standard paclitaxel dosing every 3 weeks for 6 cycles. The other arm is placebo, paclitaxel, carboplatin. Then there was a long maintenance of dostarlimab; they were able at the maintenance phase to switch to [dosing every 6 weeks], so a higher dose given less frequently. The other group was getting a placebo, every 6 weeks up to 3 years, and then they followed up. So that’s the RUBY design. Baseline patient characteristics [showed] median age very similar [to NRG-GY018]. Racial and demographic breakdown were also similar. They did not include a performance status of 2 in the study. Most patients were a performance status of 0, but it’s a pretty even split between 0 and 1. All stages were represented, but the majority were advanced stage. There were plenty of stage IV and recurrent patients included in the study. In fact, that was the majority. There was 1 patient with serous carcinoma. MMR/MSI status was reported, and most people did not have prior radiotherapy, but it made up about 15% to 20% of the total population.

What was the efficacy on the RUBY trial?

There was similar PFS in the overall population, so this is not stratified. We saw [a 24 month PFS rate] in the dostarlimab group vs placebo of 36% vs 18%, respectively [HR, 0.64; 95% CI, 0.51-0.80; P < .001]. Unsurprisingly, in the dMMR group or the MSI-high group, it was even more robust: 61.4% for the 24-month PFS rate compared with 15.7%, respectively [HR, 0.28; 95% CI, 0.16-0.50; P < .001].

When you just look at the proficient MMR population, we still see a difference: 28.4% vs 18.8%, respectively [HR, 0.76; 95% CI, 0.59-0.98]. Those CIs are not crossing 1.00. Definitely less likely to recur in those given time parameters. The AEs have a lot of overlap—fatigue, alopecia, nausea, neuropathy, anemia—so the same top 5 events that we saw in NRG-GY018. The grade 3 or more events were mostly related to blood counts. There was hypertension reported and hypokalemia. The serious events were quite rare.

Why did the investigators choose a 3-year maintenance therapy?

I haven’t seen a lot of immunotherapy maintenance studies that go out to 3 years. My personal experience is most patients don’t want to stop that if it’s working. So sometimes looking at different lengths of maintenance may be a more patient-centered approach, but [there are] also plenty of patients who can’t wait to be done. I’ve had 90-year-old patients do so well on checkpoint inhibition. [On the other hand], I just had a 22-year-old with cervical cancer who could not stay on their medication, had colitis, was on prednisone [for a long time]. We could not restart the medication safely. It is unpredictable who is going to have these AEs, and I think that’s part of the frustration. It’s hard to prepare patients.

How was the toxicity profile for patients on dostarlimab on the GARNET trial (NCT02715284)?

In terms of immune-related AEs, [it was] what you’d expect....We’re looking at more data surrounding immune-mediated responses, so hypothyroidism and hyperthyroidism are pretty consistently the most common.³ Few were grade 3 or 4 AEs in this study. Other phase 1 safety data—again, looking at patients with this dostarlimab, most treatment-related AEs were grade 1 or 2.

You don’t want to see any AEs, but it is meaningful when you’re seeing grade 1 or 2 AEs. You’re not having to stop or delay therapies. So 8.6% of patients discontinued treatment because of treatment-related AEs, and there were no deaths associated with the dostarlimab in these cohorts. This is a phase 1 trial, primarily of safety end points. Treatment-related AEs that were more common, meaning more than 10% of patients, were fatigue, diarrhea, nausea, changes in taste. Then rare grade 3 events were what you’ve seen in other studies, so you can have the elevated liver function tests, rare to have fatigue, but you can have lipase go up, early pancreatitis, hyperglycemia. There were similar rates in the dMMR and pMMR cohorts. We’re seeing even distribution in the AE profile.

CASE UPDATE

• Four months after dostarlimab initiation, partial response was noted on CT.
• During cycle 9, the patient reported fatigue, weakness, and a 10-lb weight gain.

CLINICAL WORKUP

• Free T4: low; thyroid-stimulating hormone low
• Morning cortisol and adrenocorticotropic hormone: normal
• MRI brain: normal pituitary
• Cardiac evaluation with ECG and ECHO: negative

TREATMENT

• The patient was treated with levothyroxine.
• Dostarlimab was restarted after free T4 returned to within normal limits
  

_REFERENCES

_{1. Eskander RN, Sill MW, Beffa L, et al. Pembrolizumab plus chemotherapy in advanced endometrial cancer. N Engl J Med. 2023;388(23):2159-2170. doi:10.1056/NEJMoa2302312}

_{2. Mirza MR, Chase DM, Slomovitz BM, et al; RUBY Investigators. Dostarlimab for primary advanced or recurrent endometrial cancer. N Engl J Med. 2023;388(23):2145-2158. doi:10.1056/NEJMoa2216334}

_{3. Oaknin A, Gilbert L, Tinker AV, et al. Safety and antitumor activity of dostarlimab in patients with advanced or recurrent DNA mismatch repair deficient/microsatellite instability-high (dMMR/MSI-H) or proficient/stable (MMRp/MSS) endometrial cancer: interim results from GARNET-a phase I, single-arm study. J Immunother Cancer. 2022;10(1):e003777. doi:10.1136/jitc-2021-00377}