Is there a better way to treat ER/PR+, HER2- localized breast cancer?
Data from three trials presented at the end of last year helped me try to answer this question for myself.
Trial #1: KEYNOTE-756
ITT Population:
- T1c-2(> 2 cm), cN1-N2 or T3-T4, N0-N2
- Grade III
- Invasive Ductal Carcinoma ER/PR+, HER2-
- Median age of 49
- PDL1 CPS 1 or greater in 75% and > 10 in 40% of immunotherapy group
- 2/3 were Stage II
- 80-90% were lymph node +
Study Arms:
- Neoadjuvant chemoimmunotherapy with pembrolizumab q3 weeks with paclitaxel weekly for 12weeks followed by 4 cycles of AC vs neoadjuvant chemotherapy + pembrolizumab
- After surgery, pembrolizumab for 9 cycles q3 weeks + endocrine therapy
Primary Endpoints:
- Pathological complete response and event free survival
Results
- 1278 total patients, roughly 50/50 each arm
- At first interim analysis with median follow up of 33 months
- ITT population: statistical improvement in pCR with chemoimmunotherapy 24% vs 16%with chemotherapy alone
- Most side effects in immunotherapy arm were thyroid related at 32% with any gradeand 7% with grade 3-5 adverse effects.
What to think?
This is a potentially practice changing finding. Chemoimmunotherapy is commonly being utilized in the neoadjuvant setting with lung being one of the first to develop such similar schemas to downstage patients prior to surgery followed by triple negative breast cancer in KEYNOTE 355. Traditionally neoadjuvant chemotherapy in the ER/PR+ setting has had poor efficacy with pCR rates of under 10% typically regardless of lymph node status and those who respond are typically higher grade (2/3) with slightly different biology compared to grade1 or 2 individuals. Response seems to correlate with higher PDL1 expression with 1% being where benefit starts to be shown the most. Most side effects were immunotherapy related. This is the first interim analysis but keep a close eye on this one as this could change our practice for this difficult population of patients to treat very quickly with longer term follow up!
Trial #2: Checkmate 7FL
ITT Population
- T1c-2(> 2 cm), cN1-N2 or T3-T4, N0-N2
- Grade II with ER 1-10% and Grade III with ER 1% or greater
- Invasive Ductal Carcinoma ER/PR+, HER2-
- Median age of 50
- PDL1 CPS 1 or greater in 75% and > 10 in 40% of immunotherapy group
- Equal split between Stage II and III
- Ki67>20% in 42% of chemoimmunotherapy arm
- Axillary nodes positive in 80% of chemoimmunotherapy arm
Arms
- Neoadjuvant chemoimmunotherapy with nivolumab q3 weeks with paclitaxel weekly for 12 weeks followed by 4 cycles of AC (q2 vs q3 wks) + nivolumab (360 q3 wks or 240 q2wks) vs neoadjuvant chemotherapy + placebo
- After surgery, nivolumab for 7 cycles q4 weeks + endocrine therapy
Primary Endpoints:
- Pathological complete response and event free survival
Results
- Long term follow up at 12 months post surgery
- At first interim analysis with median follow up of 33 months
- ITTpopulation: statistical improvement in pCR with chemoimmunotherapy 24% vs 16%with chemotherapy alone
- Mostside effects in immunotherapy arm were thyroid related at 32% with any gradeand 7% with grade 3-5 adverse effects.
- Thosewith increasing sTILs > 1% had greatest benefit from nivolumab
- Grade3, highest pCR from nivolumab in ER < 50% and ER < 10% with pCR of 51 and55%.
- Grade3, pCR rates of 40% in PR < 10% and in grade 3 ER >10% but PR < 10% pCRof 35%
- No association between Ki-67 and nivolumab benefit
What to think?
First thing is that these findings are all in an exploratory analysis which should betaken with a grain of salt. But a lot of interesting questions I found myself asking throughout this presentation at SABCS. Should sTILs be used as a potential biomarker in other settings and correlate well to PDL1 and immune response? I think one of the most striking things about this analysis was that the grade 3 ER < 50% and < 10% groups had the highest pCR rates up to 50%range along with those with PR < 10%, which points a different biology in these patients and what we truly classify as ER+ and PR+? There is also a notion that was challenged here that those with a higher Ki67 percentage based on adjuvant studies like NATALEE and others that a higher Ki67 percentage are more likely to respond to chemotherapy and/or immunotherapy. Interesting findings overall.
Trial #3: Phase IIIINVAVO120
ITT population
- 325 patients with locally advanced/metastatic ER/PR+, HER2-, PIK3CA mutant breast cancer
- 60% white
- No prior systemic therapy for their cancer
- Could have had relapse within/during 12 months of adjuvant endocrine therapy
Arms
- Triplet therapy of inavolisib 9 mg daily + palbociclib 125 mgdays 1-21 + fulvestrant versus placebo + palbociclib + fulvestrant
Primary endpoint
- PFS
Results
- At median follow up of 21 months, 42% of experimental arm still on treatment
- Significant PFS with median PFS 15 months in experimental arm vs 7.3 months in control, HR0.43.
- Median OS not reached in experimental arm vs 31 months in control
- ORR: 58% in experimental arm vs 25% in control arm
- DOR: 18 months in experimental arm vs 9.6 months, HR of 0.57
- G3/G4 overall safety (AEs) of both arms
- Neutropenia (80% vs 78%), thrombocytopenia (14% vs 4%)
- G3/G4 in experimental arm
- Hyperglycemia(6%), diarrhea (4%), stomatitis/mucosal inflammation (6%)
What to think?
This is the one of the first triplet combination tested in frontline metastatic PIK3CA hormone receptor positive metastatic breast cancer, which makes it a very exciting study and option for this subgroup of patients. Fairly significant difference in PFS at current follow up with long duration of response and high overall response rate. The thing that stood out to me is that unlike other PIK3CA inhibitors have been poorly tolerated by patients- the rate of serious G3/G4 adverse events with hyperglycemia, diarrhea, and stomatitis with this combination were relatively low. We will need a further look into side effect profile for G1/G2 to confirm this as well. With short term follow up, I don’t think we can draw any conclusions from here yet, but very promising concept and study overall.
Another trial that I found particularly interesting was MONARCH3. The overall survival analysis with 8 year follow up evaluating abemaciclib + aromatase inhibitor in those with metastatic ER/PR+, HER2- breast cancer did not show statistical significance unfortunately, but did show a numerical difference of 67 months for experimental arm vs 54 months in AI alone arm. Within subgroup analysis, the OS benefit was more pronounced in those with prior AI therapy and progesterone receptor negative status interestingly. There was no statistical OS difference in those with visceral disease (but again number 64 months in experimental arm vs 49 months in control arm). The updated PFS in the ITT population did show a statistical improvement in PFS from 29 months in experimental arm vs 15 months in control arm (HR 0.54).
