I-SPY2.2: A Personalized Approach to Neoadjuvant Breast Cancer Treatment
Rebecca A. Shatsky, MD, discussed the I-SPY2.2 trial, its novel design, and its implications for the treatment of patients with breast cancer.
Breast cancer cells: © LASZLO- stock.adobe.com
As breast cancer treatment continues to evolve, novel therapies and combinations move to the forefront to be investigated for curative intent.
I-SPY2.2 (NCT01042379) is a clinical trial designed to evaluate novel experimental treatments in breast cancer. The study utilizes a sequential multiple assignment randomized trial (SMART) design, which allows patients to receive multiple treatments based on their individual tumor response.
The trial uses a tumor response predictive subtype to categorize patients into 6 groups based on their tumor characteristics. This allows for tailored treatment plans.
In treatment block A, patients initially receive a novel experimental treatment of the antibody-drug conjugate datopotamab deruxtecan (Dato-DXd) and durvalumab (Imfinzi). If necessary, patients may proceed to standard chemotherapy and targeted therapies in blocks B and C.
The primary end point of the study is achieving a pathological complete response. The trial uses Bayesian models to estimate the efficacy of the treatments and to determine if they meet the threshold for further investigation.
In the Dato-DXd and durvalumab arm, 106 patients were enrolled, and 36 proceeded to surgery without completing the full treatment regimen. Dato-DXd plus durvalumab showed promising results in the HER2-immune responsive subtype and may warrant further investigation in a larger trial.
Overall, I-SPY2.2 is a valuable study that is helping to advance personalized treatment options for breast cancer patients. By using a SMART design and focusing on tumor response, the trial has the potential to identify effective new treatments for this disease.
In an interview with Targeted OncologyTM, Rebecca A. Shatsky, MD, I-SPY2.2 investigator and breast medical oncologist at the University of California San Diego Health, discussed the trial, its novel design, and its implications for the treatment of patients with breast cancer.
Targeted Oncology: Could you summarize some of the previous research on Dato-DXd in breast cancer?
Rebecca A. Shatsky, MD
Shatsky: Dato-DXd was examined in breast cancer in the TROPION-Breast trials, which showed promising efficacy in the heavily pretreated breast cancer setting, and [that was the] reason that we were excited about using Dato-DXd and durvalumab together in breast cancer in the neoadjuvant setting.
I-SPY was based on the BEGONIA trial (NCT03742102), which was a trial for first-line metastatic triple-negative breast cancer. The trial had patients who had not received a taxane in the last 12 months, and it showed promising efficacy with an overall objective response rate of 79% in that setting.
We know that topoisomerase I inhibitors potentially can enhance antitumor efficacy and enhance the efficacy of immunotherapy drugs, so the combination of combining an immune checkpoint inhibitor and this topoisomerase I inhibitor in an ADC was promising to move as soon as we could into the curative setting.
What are some of the unmet needs in the patient population?
The I-SPY2.2 trial focuses on [patients] with high-risk breast cancer. What I mean by that is that all patients must have a molecular assay that says that their breast cancer subtype would respond beneficially from the addition of chemotherapy. They had to be MammaPrint [assay] high-risk and stage II or III breast cancer. That patient population excludes the very indolent estrogen-positive subtypes that have low risk on their MammaPrint assays. Even if they have multiple lymph nodes positive, we are really focused here on high molecular risk patients with early-stage breast cancer in the curable setting.
What were you evaluating in the study?
This is not just a trial that looks at the efficacy of different medications for breast cancer. We have a section where we develop new and exciting biomarkers to reclassify breast cancer into more specific subtypes that are both prognostic but also predictive of a patient's potential response to therapy. We call those response predictive subtypes, and those are I–SPY-specific biomarkers.
The trial has arms where we focus on pathology. We also have parts where we focus on radiology. We have developed breast MRI functional tumor volume algorithms with our team specifically that we use and incorporate to evaluate response within the trial. Then we also look at the efficacy of drugs in on their own in the first part of the trial as well as looking at novel investigative agents in combination with standard-of-care therapy for patients who need more than just the investigational agents that we are utilizing upfront.
What is great about it for patients is that the primary end point is pathologic complete response, but we try our very best to maximize that end point for each individual patient. So, we know that not all patients are going to be able to achieve a pathologic complete response with just the first part of therapy of investigational agents, and that is why if we do not predict that [those agents] are going to [deliver] a pathologic complete response for them, we encourage them to go on to the second and third parts of the trial, which then give them the best chance of maximizing their own individual chance of getting a pathologic complete response since we know that that is prognostic for their overall survival in breast cancer.
Could you summarize your efficacy findings?
The efficacy findings in in I-SPY2.2, the newest version of the trial, are always a little bit complicated because again, we are not allowing every patient that goes on trial to immediately go to surgery early after the upfront combination therapy. For this trial, the novel agents that we investigated were 4 cycles of datapotamab deruxtecan combined with 4 cycles of durvalumab. For those patients that we predicted would have benefit, they went on, and about 33% of all patients who went on this arm of the trial—there was 106 patients—33% were able to just do 4 cycles of datapotamab deruxtecan and durvalumab and skip traditional chemotherapy, which includes skipping the toxicities of a taxane-based therapy, meaning neuropathy and the potential use of an anthracycline, which could cause leukemia or cardiotoxicity.
For [patients with triple-negative breast cancer], on this first phase of the trial, in our most conservative estimate, we saw about a 33% chance of pathologic complete response. However, I-SPY utilizes a complicated Bayesian covariant analysis modeling system where we could model if we had allowed all patients to go to surgery after block A of datapotamab deruxtecan and durvalumab alone, what percentage of patients would have achieved a pathologic complete response if we allowed them to? For [patients with triple-negative breast cancer], that was 43%.
For the subgroup that we have identified by our own biomarkers, which is our immune responsive subgroup, and that includes both [patients with] hormone-positive and triple-negative [disease] that had this immune biomarker that suggests sensitivity to immunotherapy, what we modeled in that group is that their pathologic complete response rate would be 65%. We find that valuable data because we have identified this subgroup of [patients] who we can de-escalate therapy on to avoid potential toxicities by over treating them. But at the same time, of course, in the trial, we do not want to under treat anybody, so we allow patients, if they are not likely to achieve a pathologic complete response, to move on to additional blocks of therapy based on the standard of care.
For this combination, what adverse effects have been observed or associated with this combination? How can clinicians mitigate these toxicities?
On the forefront of every patient and clinician’s mind is, what toxicities is this drug combination going to have? For datapotamab plus durvalumab, what we see is that stomatitis is common. That has been consistent amongst all the trials of this drug, but it is a real toxicity in our trial. We provided patients with dexamethasone mouthwash to try to mitigate that. There definitely does seem to be a positive benefit to utilizing dexamethasone mouthwash for patients receiving datapotamab. The compliance on our trial of the use of dexamethasone mouthwash was not super closely tracked, so it could be variable. But that is probably the most common toxicity of that drug.
The other things that we saw that were important and unique to this combination [included] rash. There was a decently high rate of rash, but it did resolve very quickly with either topical steroids or antihistamines, and it was not severe. It was not grade 3 or 4 events that we were seeing. Nausea is present, but it is on the milder side, so grade 1 and 2.
Overall, in this trial, the grade 3 and 4 toxicity event rates were pretty low. Importantly, with this combination, we did not utilize the addition of growth factors to prevent neutropenic fever. There were only 2 events in this arm of grade 3 neutropenia. That was helpful because many of the other drugs use in the early breast cancer setting have high rates of severe neutropenia and high rates of neutropenic fever, and the same thing for anemia. We saw a very mild anemia here and not necessitating transfusion in patients, so that is an important positive finding.
What are the next steps in researching this combination or in the I-SPY trial?
We are excited about this combination in our immune responsive subtype. We think that, based on our mathematical modeling, this group would have a greater than 85% chance of having success in a phase 3 trial. We would encourage further research in phase 3 clinical trials in the curative setting of this combination for specific subtypes of patients.
As far as I-SPY, we continue to add more sites around the country. We are a very large clinical trial organization, and it is a grassroots effort. We partner with pharmaceutical companies, but a lot of these different investigational algorithms are developed by our own internal team. In addition, we keep adding new novel subgroups and drug combinations to see what may be the most effective in the most select population so that we can individualize therapy and move promising drug combinations into the curative setting earlier than they would have been investigated otherwise.
