Cyclin-dependent kinase 4/6 inhibitors, known as CDK4/6 inhibitors, work by blocking two enzymes that drive the cell division cycle. In hormone receptor-positive breast cancer, tumor cells rely heavily on signaling through cyclin D and the kinases CDK4 and CDK6 to proliferate, and pairing one of these inhibitors with endocrine therapy such as an aromatase inhibitor or fulvestrant has been shown in landmark phase III trials—including PALOMA, MONALEESA, MONARCH, and DAWNA—to dramatically extend the time patients live without their disease progressing. Yet pivotal clinical trials enroll carefully selected patients under tightly controlled conditions, and the outcomes of ordinary patients in routine clinical practice, who are often older, have more comorbidities, or fall outside trial eligibility criteria, can differ substantially. That gap between trial efficacy and real-world effectiveness is precisely what the new study was designed to address.

The retrospective multicenter analysis drew on records from patients treated at 20 cancer centers across central China, making it one of the most geographically diverse real-world datasets of its kind. CDK4/6 inhibitors were used as first-line therapy in 65.68 percent of the cohort and as second-line treatment in 24.60 percent, with the remainder receiving the drugs later in their treatment course. The primary endpoint was progression-free survival, the length of time a patient lives without evidence of tumor growth or spread, assessed using imaging criteria and Kaplan–Meier statistical methods. The findings confirmed a striking advantage for earlier use: median progression-free survival reached 38.0 months in patients who received a CDK4/6 inhibitor as their first systemic treatment for metastatic disease, compared with 18.8 months among those who began the drugs only after prior lines of therapy had failed, a difference that was highly statistically significant with a P value below 0.001. In other words, patients who received the drugs first lived roughly twice as long without progression.

Beyond treatment timing, the investigators used multivariable Cox regression analysis to identify which patient characteristics independently shaped prognosis. Cox regression is a statistical technique that estimates the effect of multiple variables simultaneously on the risk of an event such as disease progression, while accounting for the fact that not all patients have been followed for the same length of time. Three factors emerged as adverse prognostic markers: having the Luminal B molecular subtype of breast cancer, which tends to be more aggressive than Luminal A disease; the presence of liver metastases, a known indicator of higher disease burden; and receiving the CDK4/6 inhibitor as second-line rather than first-line treatment. Conversely, two features were associated with better outcomes: tumors with HER2 immunohistochemistry score of 1+, a faint level of HER2 protein expression sometimes called HER2-low, and a longer disease-free interval between the initial diagnosis and the development of metastatic disease. Each of these findings aligns with, and extends, signals from smaller studies conducted in Europe, Japan, and North America.

To translate these population-level findings into a tool usable at the bedside, the team split patients receiving first- or second-line CDK4/6 inhibitors into a training cohort and a validation cohort in a seven-to-three ratio. On the training data they built a conventional Cox regression model and seven distinct machine learning algorithms designed for survival data: gradient boosting machines (GBM), random survival forests (RSF), Lasso-Cox, CoxBoost, XGBoost, super principal component analysis (SuperPC), and partial least squares regression for Cox data (plsRcox). These methods differ in how they handle complexity. Random survival forests, for example, grow many decision trees on bootstrap samples of the data and average them to capture non-linear relationships, while gradient boosting builds an ensemble of weak learners sequentially, each correcting the errors of the last. Lasso-Cox applies a penalty that shrinks coefficients and performs variable selection automatically, guarding against overfitting in datasets with many correlated predictors.

Model performance was evaluated using three complementary approaches: time-dependent area under the receiver operating characteristic curve (AUC), which measures discrimination, meaning the ability to correctly rank patients who progress sooner above those who progress later; calibration plots, which test whether predicted probabilities match observed outcomes; and decision curve analysis, which quantifies the clinical net benefit of acting on the model’s predictions at various risk thresholds. The conventional Cox model achieved respectable discrimination, with AUCs of 0.731, 0.719, and 0.704, values that indicate clinically meaningful predictive accuracy without reaching the level of certainty that would justify replacing clinician judgment. Among the machine learning approaches, gradient boosting machines and random survival forests showed the highest discrimination in the training cohort but settled into only moderate performance when tested on the held-out validation cohort, a pattern that reflects the classic challenge of overfitting, in which flexible algorithms memorize quirks of the training data that do not generalize to new patients.

The comparison between the Cox model and the machine learning alternatives carries a broader lesson for the field of computational oncology. Machine learning methods are often assumed to outperform classical regression simply because they are more sophisticated, but the evidence from survival prediction research is mixed, and recent systematic reviews have found that the two approaches frequently perform comparably when applied to modest-sized clinical datasets. The authors of the new study conclude that both the Cox model and the machine learning frameworks enable individualized prognostic prediction for CDK4/6 inhibitor therapy, but they emphasize that the GBM and RSF models performed relatively better and that external validation in independent patient populations remains essential before any of the tools can be deployed in routine clinical practice. This cautious stance mirrors the standards set by the TRIPOD reporting guidelines, which require transparent documentation of prediction model development and validation.

The study’s real-world effectiveness data carry important implications for treatment sequencing guidelines. Because median progression-free survival was double in the first-line setting, the findings reinforce the strategy of deploying CDK4/6 inhibitors upfront in combination with endocrine therapy rather than reserving them for later lines, consistent with the design of trials such as PALOMA-2, MONALEESA-2, MONARCH 3, and DAWNA-2. The finding that HER2-low tumors fared better adds to a growing body of evidence that the HER2-low subgroup, which was historically lumped together with HER2-zero disease, may represent a biologically and clinically distinct entity, with consequences for eligibility for novel antibody-drug conjugates as well. Meanwhile, the adverse prognostic weight of liver metastases and Luminal B biology provides clinicians with concrete variables to weigh when counseling patients and planning surveillance intensity.

The research also has significance for Chinese and other Asian patient populations, where locally relevant real-world evidence has historically been thinner than in Western Europe and North America. The cohort included patients treated with agents available in China, and the treatment patterns observed—first-line use in roughly two-thirds of patients—suggest substantial but incomplete uptake of guideline-concordant sequencing. The study protocol was registered at ClinicalTrials.gov, conducted under the Declaration of Helsinki, and approved by the Ethics Committee of Hunan Cancer Hospital, which waived the requirement for individual written informed consent given the retrospective, anonymized nature of the data. Funding came from the Hunan Provincial Natural Science Foundation, Hunan Cancer Hospital programs, and two Chinese medical foundations, and the authors declared no competing interests.

For patients with hormone receptor-positive, HER2-negative metastatic breast cancer, the most immediate message is one of cautious optimism: in the messy reality of everyday oncology, CDK4/6 inhibitors deliver substantial benefit, with first-line patients in this large cohort living a median of more than three years without progression. For the oncology community, the study demonstrates a rigorous template for building prognostic tools from real-world data, combining the interpretability of classical survival regression with the flexibility of modern machine learning. And for the rapidly expanding field of AI-assisted medicine, it serves as a measured reminder that predictive power must be validated, calibrated, and externally confirmed before an algorithm earns a place in the clinic. As external validation cohorts are assembled, the models described in this work may eventually help oncologists answer one of the most practical questions in metastatic breast cancer care: which patient, with which tumor, is likely to benefit most, and for how long, from these transformative drugs.

The study, called EVERGREEN, analyzed outcomes for 207 women whose estrogen receptor-positive, human epidermal growth factor receptor 2-negative advanced breast cancer had progressed after treatment with a CDK4/6 inhibitor. Of these patients, 150 received everolimus alongside endocrine therapy, while 57 received endocrine therapy without everolimus. After a median follow-up of 31.8 months, the median real-world progression-free survival was 5.0 months in the everolimus group, compared with 4.3 months among those given endocrine therapy alone. The adjusted hazard ratio for progression or death was 0.68, with a 95 percent confidence interval of 0.47 to 0.99.

That result means the everolimus-containing treatment was associated with an approximately 32 percent lower relative risk of progression or death during the study period after statistical adjustment. It does not mean that every patient gained a fixed additional 32 percent of survival, nor that the cancer was controlled for 32 percent longer. The absolute difference in median progression-free survival was only 0.7 months—roughly three weeks. The researchers therefore describe the benefit as modest. There was no statistically significant improvement in the time until chemotherapy was needed or in overall survival, the measure that most directly captures whether treatment helps patients live longer.

Everolimus attacks a different part of the machinery that cancer cells use to grow. It inhibits mammalian target of rapamycin, or mTOR, a central signaling protein that helps regulate protein production, cell growth, metabolism and survival. In hormone receptor-positive breast cancer, signaling through the estrogen receptor can cooperate with the PI3K–AKT–mTOR pathway to keep malignant cells dividing even when estrogen-driven growth is being suppressed. Laboratory studies have suggested that increased activity in this pathway can contribute to resistance against CDK4/6 inhibition. By blocking mTOR while continuing endocrine therapy, clinicians aim to shut down a bypass route that cancer cells may exploit after cell-cycle treatment stops working.

The biological rationale, however, does not guarantee a large clinical effect. Tumors that acquire resistance to CDK4/6 inhibitors are not uniform. Some develop alterations in genes such as ESR1, which encodes the estrogen receptor; others involve the PI3K–AKT–mTOR network, including PIK3CA, AKT1 or PTEN. Still others may become less dependent on estrogen signaling altogether, switch to alternative growth programs or contain several resistant subclones at once. Everolimus may be most useful when the mTOR pathway remains an important engine of tumor growth, but the EVERGREEN study did not establish a biomarker that could reliably identify such patients before treatment.

The study’s design is important for interpreting its findings. EVERGREEN was a multicentre, international, retrospective quasi-experimental study rather than a randomized clinical trial. The investigators compared women treated at centers where everolimus plus endocrine therapy was the standard approach with women treated at centers where endocrine therapy alone was standard. This approach can provide valuable evidence from routine oncology practice, especially when randomized trials have not answered a specific treatment question. It also introduces potential sources of bias: treatment policies differ between hospitals, physicians may select everolimus for particular types of patients, and medical records may not capture every factor influencing treatment choice or disease assessment.

The patient groups were broadly balanced at baseline, according to the researchers, but the everolimus cohort had received a greater number of previous lines of therapy. That imbalance matters because heavily pretreated disease can be more biologically resistant and patients may have poorer overall health or fewer remaining treatment options. The investigators used adjusted analyses to account for measured differences, but statistical methods cannot completely remove the effects of unknown or unrecorded factors. “Real-world progression-free survival” is also less tightly controlled than progression-free survival in a prospective trial, where imaging schedules, response assessments and follow-up procedures are standardized.

The safety findings were consistent with earlier reports of everolimus, but the abstract does not provide a detailed breakdown of adverse events. The drug can cause mouth inflammation, rash, fatigue, diarrhea, metabolic changes and suppression of blood-cell production; it can also produce noninfectious pneumonitis, an inflammatory lung complication. These risks are particularly relevant in advanced cancer, where maintaining quality of life is a central treatment goal. A therapy that delays progression by several weeks may be worthwhile for a carefully selected patient who wants to remain on oral treatment and has limited alternatives, but less attractive for someone vulnerable to complications or eligible for a better-matched molecular therapy.

The findings arrive in a rapidly changing treatment landscape. After CDK4/6 inhibitor progression, options may include endocrine drugs designed to target specific resistance mutations, inhibitors of PI3K or AKT signaling, antibody–drug conjugates and chemotherapy. For example, the presence of an ESR1 mutation or an alteration in PIK3CA, AKT1 or PTEN may influence the suitability of other targeted approaches, although the best sequence of therapies is still evolving. The EVERGREEN results do not show that everolimus should replace these options. Instead, they suggest that it remains a possible strategy for a subset of patients whose disease is still endocrine-sensitive and for whom the expected benefits outweigh the drug’s side effects.

The study also illustrates why treatment after CDK4/6 resistance cannot be reduced to a single universal prescription. A median benefit measured in weeks can conceal meaningful differences between individuals: some patients may experience little response, while others may achieve substantially longer disease control. Future trials will need to connect outcomes to tumor biology, circulating tumor DNA, prior endocrine exposure and the pattern of progression. For now, EVERGREEN provides a cautiously encouraging but not practice-revolutionizing message: everolimus can add a small amount of control after CDK4/6 inhibitors, but the decision to use it should be individualized rather than automatic.

The phase 2 study evaluated cemiplimab and fianlimab alongside standard neoadjuvant chemotherapy. Neoadjuvant treatment is delivered before surgery, allowing physicians to measure how effectively a therapy eliminates cancer from the breast and nearby lymph nodes. Patients received weekly paclitaxel followed by doxorubicin and cyclophosphamide, while the immunotherapies were administered every three weeks. Surgery was performed after the drug treatment. The trial included adults with stage II or III disease considered at high risk of recurrence, including people with triple-negative breast cancer and those with hormone receptor–positive, HER2-negative tumors.

HER2-negative breast cancer is also described as ERBB2-negative breast cancer. ERBB2 is the gene that encodes the HER2 protein, a receptor involved in cell-growth signaling. When HER2 is overproduced, tumors can often be treated with HER2-directed drugs; when it is not, treatment generally relies on chemotherapy, endocrine therapy when hormone receptors are present, and increasingly, immunotherapy for selected patients. The I-SPY2 study was designed to identify promising combinations quickly by comparing investigational regimens with standard therapy and using early results to estimate whether a treatment might succeed in a larger phase 3 trial.

In the experimental arm, cemiplimab blocked PD-1, an inhibitory receptor on T cells. Tumors can exploit the PD-1 pathway to weaken immune-cell activity and avoid destruction. Fianlimab targeted LAG-3, another immune checkpoint that can suppress T-cell function when persistently activated. Blocking both pathways was intended to release complementary brakes on the immune response, potentially allowing immune cells to recognize and attack malignant cells more effectively than either approach alone. A total of 78 patients received the combination, while 350 participants were assigned to the control group.

The primary endpoint was pathologic complete response, or pCR, defined as the absence of residual invasive cancer in the breast and lymph nodes at surgery. Although pCR is not identical to long-term cure, it is strongly associated with a lower risk of recurrence in many high-risk breast cancers and is widely used to assess the effectiveness of preoperative treatment. Across all HER2-negative participants, the estimated pCR rate rose from 21% with standard therapy to 44% with the cemiplimab-fianlimab regimen. The result exceeded the I-SPY2 platform’s prespecified threshold for predicted success in a future phase 3 study.

The apparent benefit was observed in both major disease subgroups. Among patients with triple-negative breast cancer, the estimated pCR rate increased from 29% with standard therapy to 53% with the dual-immunotherapy regimen. In hormone receptor–positive, HER2-negative disease, the rate rose from 14% to 36%. These results suggest that combined checkpoint inhibition may have activity beyond tumors traditionally considered most responsive to immunotherapy. However, the study was not designed to establish definitive survival advantages, and the investigators emphasized that the exact regimen should not be adopted as a new standard based on these findings alone.

Tumor biology appeared to influence the magnitude of response. Patients whose tumors were positive for ImPrint, a 53-gene test designed to identify an immune-responsive tumor environment, experienced particularly high pCR rates after receiving the experimental treatment. The estimated rate reached 83% among ImPrint-positive patients with triple-negative disease and 91% among those with hormone receptor–positive, HER2-negative tumors. ImPrint-positive tumors show gene-expression patterns associated with immune-cell activity and a microenvironment that may be more receptive to immune stimulation. The findings support the broader goal of matching immunotherapy to the molecular features of each patient’s tumor.

The safety results nevertheless placed important limits on the treatment’s future development. Decreased adrenal hormone production occurred in 21% of patients receiving the dual-checkpoint combination, including grade 3 or 4 events in 11%. Diabetes developed in 4% of participants. These endocrine toxicities can require long-term hormone replacement or specialist care, and some emerged weeks after immunotherapy had ended. The delayed timing illustrates why patients receiving checkpoint inhibitors need continued monitoring after treatment, even when chemotherapy and surgery have been completed. The investigators concluded that the efficacy signal was compelling but that the toxicity profile was too concerning for the regimen to advance unchanged.

Claudine Isaacs, MD, lead author and associate director for clinical research at Georgetown’s Lombardi Comprehensive Cancer Center, said the results establish a proof of principle for targeting two immune checkpoints at once. Future studies may examine different doses, schedules, or newer drugs designed to hit both pathways with a potentially lower toxicity burden. One possibility is the use of bispecific antibodies, engineered molecules that bind two targets through a single drug. The I-SPY2 platform, which has enrolled more than 2,500 patients and tested 25 therapies over approximately 15 years, will continue evaluating treatment combinations and molecular tests intended to maximize benefit while limiting unnecessary harm.

Subject of Research: People

Article Title: Cemiplimab and Fianlimab With Neoadjuvant Chemotherapy in Early-Stage High-Risk ERBB2-Negative Breast Cancer: The I-SPY2 Randomized Clinical Trial

News Publication Date: 30-Jul-2026

Web References: https://doi.org/10.1001/jamaoncol.2026.2576; ClinicalTrials.gov identifier NCT01042379

References: JAMA Oncology, DOI: 10.1001/jamaoncol.2026.2576

Keywords: breast cancer, HER2-negative breast cancer, ERBB2, immunotherapy, cemiplimab, fianlimab, PD-1, LAG-3, neoadjuvant chemotherapy, pathologic complete response, I-SPY2, precision medicine