A new breast cancer classification could sharpen immunotherapy targeting

Introduction: when science finally refines its predictive tools

A problem as old as immunotherapy itself

Since immune checkpoint inhibitors (ICIs) arrived in the therapeutic arsenal against breast cancer, one problem has persisted, frustrating patients and oncologists alike: a significant share of treated people simply don't respond to these drugs, despite their spectacular initial promise. This clinical reality, documented for years, has pushed research teams around the world to look for better tools to identify, before treatment even begins, who will actually benefit.

A study published in 2026 in the journal Cancer Biology & Medicine, led by researchers from the Department of Breast Surgery at Fudan University Shanghai Cancer Center and Shanghai Medical College, offers precisely an answer to this question, in the form of a new classification system based on what scientists call the cancer immunity cycle, or CIC.

The cancer immunity cycle, explained simply

The CIC is a conceptual framework describing, step by step, how the immune system is supposed to recognize and destroy cancer cells: from the release of tumor antigens to the final killing of cancer cells by T cells. When any one of these steps fails, the whole process can collapse, and the tumor escapes immune surveillance.

The researchers developed a CIC score that measures the activity of six key steps of this antitumor immune response. By analyzing this score across a large number of patients, the team was able to sort breast tumors into three distinct groups, each with its own biological characteristics and its own outlook for response to immunotherapy.

Three tumor profiles, three different therapeutic destinies

Cluster C1, the cold, hard-to-treat tumor

The first group identified by the researchers, called C1, corresponds to what oncologists call an "immunologically cold"tumor. These tumors show low immune cell infiltration, a grimmer prognosis, and an abundance of immunosuppressive M2 macrophages, cells that, rather than helping fight the cancer, help create an environment favorable to its progression.

This category of patients is precisely the one that benefits least from current checkpoint inhibitors, which clinically means they could potentially be spared costly, lengthy treatments carrying significant side effects, in favor of other therapeutic approaches better suited to their specific biological profile.

Cluster C3, the hot tumor that responds well

At the opposite end of the spectrum is cluster C3, described as an "immunologically hot"tumor, characterized by heavy immune cell infiltration, active T cells, and, unsurprisingly, the best observed response to checkpoint inhibitor therapies among the three groups studied by the Chinese research team.

This distinction between "cold" and "hot"tumors isn't new in oncology, but what this study contributes is a more rigorous, quantifiable method for establishing that distinction, rather than relying on qualitative observations that are sometimes imprecise across different labs and hospitals.

Cluster C2, the study's most unexpected finding

A biological trap no one had properly mapped

It's the second cluster, dubbed C2, that the researchers consider the most surprising discovery of this study. This intermediate subtype shows a particular defect in antigen presentation, an essential mechanism through which cancer cells are normally "flagged" to the immune system for destruction.

What makes this group especially treacherous is that these tumors often display a high tumor mutational burden (TMB), an indicator that, in most cancers, usually suggests a good response to immunotherapy. Yet C2 tumors frequently defy this general rule because of a phenomenon called HLA loss of heterozygosity, which literally scrambles the signal the tumor should be sending to the immune system.

A hostile tumor microenvironment despite appearances

In addition to this antigen presentation defect, C2 group tumors show an immunosuppressive tumor microenvironment, enriched with dysfunctional dendritic cells and regulatory T cells, two cell types that, together, help smother any attempt at an effective immune response against the tumor.

This combination explains why, despite genetic features that look promising on paper, these patients don't respond as well as expected to standard treatments, a clinical paradox this new classification finally helps us better understand and, potentially, work around.

Distinct metabolic dependencies for each subtype

Sphingolipid metabolism and cluster C1

Beyond the immune classification, the researchers conducted multi-omics analyses that revealed specific metabolic dependencies for each identified cluster. The C1 group, the cold tumor, shows particular enrichment of sphingolipid metabolism, a family of lipid molecules involved in cell signaling and, potentially, immune evasion.

This information isn't just a biochemical curiosity: it opens the door to combination therapies specifically targeting these metabolic pathways in patients whose tumor belongs to this subtype, a strategy that could potentially warm up a cold tumor and make it more receptive to immunotherapy.

PSAT1 and serine metabolism in cluster C2

For cluster C2, the research team identified a strong dependence on serine metabolism, with a particularly notable role for the enzyme PSAT1, identified as a key metabolic regulator in this specific tumor subtype. Notably, suppressing this enzyme in cancer cells in the lab reduced the expression of key immunosuppressive molecules like PD-L1 and TGFB1.

This observation suggests there could be a concrete therapeutic target for trying to disarm the immune-escape mechanism specific to the C2 group, a lead the researchers explicitly identify as a future direction for their follow-up work.

What the researchers say about their own discovery

Moving past the simplistic hot-and-cold paradigm

In their own words, the study's authors say the CIC offers "a powerful framework for understanding how tumors evade the immune system." They add that by building a comprehensive score capturing the efficiency of the entire cycle, their team was able to "move beyond the simple hot-cold paradigm" to identify distinct, therapeutically exploitable defects.

This nuance matters: for years, the oncology community has tended to reduce tumor complexity to a simple dichotomy between hot and cold tumors, a useful initial simplification that, as we now see, let intermediate groups like cluster C2 slip through, potentially numerous in real clinical practice.

Direct implications for future clinical practice

The researchers stress that this new classification not only helps predict which patients will benefit from current immunotherapies, but also shows precisely where the cycle breaks down in each individual case, pointing toward more targeted combination strategies to improve outcomes for a greater number of patients.

In concrete terms, this means a CIC score could one day be used clinically to guide treatment decisions, identifying who will benefit most from checkpoint inhibitor treatment and sparing others unnecessary side effects for an uncertain clinical benefit.

The broader context of breast cancer immunotherapy

Real successes, but still limited to certain subtypes

It's worth remembering that the use of immune checkpoint inhibitors in breast cancer remains, to date, largely concentrated on triple-negative breast cancer (TNBC), a particularly aggressive subtype accounting for about 15 to 20% of cases but responsible for a disproportionate share of deaths from the disease. Landmark clinical trials like KEYNOTE-522 established the efficacy of pembrolizumab combined with chemotherapy in this specific setting.

But even in this subtype most receptive to immunotherapy, not all cases respond the same way, which is exactly why finer stratification tools, like the one proposed by the Fudan University team, are so actively sought by the international scientific community.

A global race to sharpen precision medicine

This Chinese study is part of a broader global scientific movement, in which teams in North America, Europe, and Asia are simultaneously trying to improve predictive biomarkers for immunotherapy response, whether through PD-L1 expression, counting tumor-infiltrating lymphocytes, or, now, composite scores like CIC.

This international scientific competition, far from being a problem, generally speeds up the pace of discovery, as each team tests and validates the others' hypotheses in different cohorts, strengthening the overall robustness of accumulated knowledge about this complex disease.

The methodological limits worth keeping in mind

A study that still needs validation at larger scale

As with any recent scientific discovery, these results should be approached with appropriate caution. The study was conducted on validation cohorts including pan-cancer data to test the predictive value of CIC subtypes in patients receiving checkpoint inhibitor therapy, but large-scale clinical application will still require prospective clinical trials specifically designed to validate this score's usefulness in real treatment decision-making.

This clinical validation stage generally takes several years in oncology, and it would be premature to promise patients an immediate change in their care based solely on this publication, however solid it is as basic research.

The persistent gap between lab discovery and clinical practice

The history of oncology is full of examples of promising research biomarkers that ultimately failed to meaningfully transform clinical practice, often because their practical implementation proved more complex or costly than expected, or because later clinical trials didn't confirm the initial promise seen in smaller cohorts.

This reality takes nothing away from the scientific value of the CIC cycle discovery, but it calls for measured enthusiasm around this kind of announcement, particularly for patients already living through a period of considerable uncertainty and anxiety about their diagnosis.

Why this research still deserves public attention

A step in the right direction for personalized medicine

Despite these necessary caveats, this study is a significant contribution to a rapidly evolving field, that of precision oncology. Every new stratification tool, however imperfect at first, brings the medical community closer to a day when treatments will be truly tailored to each tumor's unique biological profile rather than applied via generic protocols.

For patients with breast cancer, this gradual shift toward more personalized medicine represents tangible hope, even if it will likely still take several years before this kind of score enters routine clinical practice in hospitals worldwide.

The importance of basic research, even without immediate results

This study also serves as a reminder of a truth often forgotten by the general public: most significant medical advances result from years, even decades, of gradually accumulated basic research, rather than sudden, spectacular discoveries. The meticulous work of mapping the antitumor immune cycle fits squarely into this patient, cumulative scientific tradition.

It is this gradual accumulation of knowledge, publication after publication, that eventually produces real clinical breakthroughs, and it's important to value this foundational work even when it doesn't immediately lead to a new drug available at the pharmacy.

How the researchers built and validated this CIC score

A methodology in four major steps

The Fudan University researchers structured their study into four distinct methodological steps: building genetic signatures linked to the cancer immunity cycle, classifying patients according to the activity of six precise steps of this cycle, validating the predictive value of these subtypes in pan-cancer cohorts receiving immunotherapy, and then a multi-omics characterization to identify subtype-specific therapeutic targets.

This four-stage methodological rigor sets this study apart from more preliminary research that sometimes relies on simple statistical correlation without cross-validation, a technical distinction that matters enormously when assessing the real solidity of a scientific discovery in oncology.

Cross-validation across multiple cancer types

Notably, the team didn't limit its validation to breast cancer alone: it tested the predictive value of CIC subtypes in pan-cancer cohorts receiving checkpoint inhibitor treatments, suggesting this conceptual framework could eventually apply beyond breast cancer alone, though that extension still needs confirmation through studies dedicated to each specific tumor type.

This pan-cancer approach, if confirmed by future independent studies, would considerably strengthen the scientific value of this classification, since a predictive tool applicable across several cancers would have a potentially much broader clinical impact than a simple breast-specific biomarker.

The international oncology community's reaction

A cautious but broadly positive reception

In the days following this study's publication, several researchers specializing in immuno-oncology praised the rigor of the methodological approach, while noting, as is standard in basic research, that independent validation studies would be needed before any widespread clinical adoption of this classification tool.

This kind of cautious reaction, far from being a rebuke, is actually the normal, healthy functioning of the scientific method: every discovery must be reproduced and confirmed by other research teams before being considered reliable enough to influence oncologists' daily clinical practice.

A Chinese contribution in a field dominated by Western research

It's worth noting that this scientific contribution is part of the broader rise of Chinese oncology research, which now publishes a growing number of influential studies in high-impact international journals, a phenomenon several observers of the scientific sector have already documented for years.

This dynamic clearly shows that medical innovation no longer respects strict geographic borders, even though questions of methodological rigor, data transparency, and reproducibility remain universal criteria that apply equally, regardless of the country of origin of published research.

The questions that remain open for future research

The challenge of standardizing the test for clinical practice

A crucial question remains for the future: how to translate this CIC score, calculated from complex genomic analyses in the lab, into a standardized diagnostic test that could be reliably and reproducibly used in ordinary hospitals worldwide, rather than only in the most advanced university research centers.

This technical standardization step has historically been one of the biggest obstacles to translating lab discoveries into routine clinical practice, a challenge the scientific community will need to solve before this score can truly change patient care at scale.

The potential cost and equitable access to this kind of test

Another question, less often asked but just as important, concerns the potential cost of such a complex genomic test and its accessibility for patients in health systems with varying resources, including in low- and middle-income countries where the burden of breast cancer keeps rising.

Without serious thought given to accessibility and cost, even the most promising scientific tool risks remaining reserved for a minority of patients with access to the best-equipped medical centers, further widening the inequalities already present in the global fight against cancer.

What this discovery reveals about the evolution of modern oncology

The shift from general medicine to precision medicine

This study perfectly illustrates the transition underway across the entire oncology field: the gradual shift from a relatively uniform therapeutic approach, based mainly on tumor type and stage, toward truly personalized medicine, accounting for the specific molecular and immune characteristics of each individual tumor.

This transition, which began decades ago with the arrival of targeted therapies, is accelerating today thanks to combined advances in genomic sequencing, artificial intelligence applied to massive biological data analysis, and a better fundamental understanding of cancer immunology.

A model that could inspire other tumor classifications

The methodological success of this classification based on the cancer immunity cycle could also inspire similar approaches for other cancer types, where similar challenges of resistance or non-response to immunotherapy persist despite therapeutic progress made over the past decade.

This logic of functional classification, based on the actual biological activity of a process rather than simple static genetic markers, represents a methodological evolution that extends well beyond breast cancer alone and could influence oncology research as a whole in the years ahead.

The other molecular classifications coexisting with the CIC score

The Fudan classification of triple-negative breast cancer

It's interesting to note that the same university, Fudan University, had already made a significant contribution to the molecular classification of triple-negative breast cancer with what the scientific community now calls the "Fudan classification," which distinguishes four main subtypes: luminal androgen receptor, immunomodulatory, basal-like immune-suppressed, and mesenchymal.

This earlier classification, now widely cited in international scientific literature, had already helped guide certain treatment choices based on identified subtype, notably the use of PARP inhibitors for immune-suppressed basal-like tumors, an approach the new CIC score complements rather than replaces.

Complementarity rather than competition between diagnostic tools

Rather than viewing these different molecular classifications as competing, the scientific community increasingly tends to see them as complementary, each shedding light on a different aspect of the complex tumor biology of breast cancer, whether genetic, immune, or metabolic.

This complementarity could eventually lead to integrated diagnostic tools, combining multiple scores and classifications to offer a far more complete and precise picture of each individual tumor's biological profile, a still-distant prospect but one increasingly considered by researchers in the field.

The growing role of multi-omics analysis in oncology

A technical revolution that makes these discoveries possible

This study simply wouldn't have been possible without the considerable technical progress made over the past decade in multi-omics analysis, which allows simultaneous examination of the genomics, transcriptomics, and metabolism of a tumor sample, offering a comprehensive view impossible to obtain with older, more limited analytical techniques.

This enhanced technical capacity, combined with the continued decline in genomic sequencing costs, largely explains the accelerating pace of discovery in molecular oncology observed in recent years, a pace that shows no sign of slowing according to experts in the field.

The computational challenges of massive data analysis

This kind of study generates considerable volumes of data, requiring sophisticated computing power and statistical methods to extract meaningful biological signals from considerable background noise, a computational challenge increasingly pushing oncology labs to work closely with bioinformatics and artificial intelligence specialists.

This convergence between molecular biology and advanced computational science represents one of the most striking trends in contemporary medical research, a trend that will likely keep intensifying as the volume of available biological data continues to grow exponentially.

Conclusion: one more step toward better-targeted immunotherapy

What this study concretely changes, and what it doesn't change yet

In summary, this new classification based on the cancer immunity cycle offers a more rigorous scientific framework for understanding why some breast cancer patients respond to immunotherapy while others don't. It identifies three distinct tumor profiles, each with different biological characteristics and therapeutic outlooks, including the unexpected discovery of the intermediate C2 group and its specific metabolic dependencies.

What this study doesn't change yet is immediate clinical practice: additional clinical trials, larger-scale validation, and probably several more years will be needed before this CIC score is integrated into standard treatment protocols in hospitals around the world.

Measured hope rather than a miracle promise

For patients and their loved ones following this kind of scientific news with an understandable mix of hope and skepticism, the takeaway message is one of real but gradual progress: science is patiently advancing toward a better understanding of this complex disease, with no miracle shortcut but with a rigor that, over time, benefits a growing number of patients around the world.

It is precisely this kind of modest but solid advance which, accumulated over the years, has produced the real gains seen in breast cancer survival over recent decades, and there's no reason to think that trend will stop with this new discovery.

By Maxime Marquette, columnist

Columnist's transparency note

Who I am and my acknowledged biases

I sign this column as a non-specialist observer of medical issues, with a strong interest in rigorous science communication. I am neither a doctor nor an oncology researcher, and my role here is to make a complex scientific publication accessible without betraying its nuance or overstating its immediate clinical significance.

My deliberate approach systematically favors caution over enthusiasm, particularly on medical topics that directly affect people living with a cancer diagnosis, a population for whom false hope can carry a real emotional cost.

What I don't know, and my method

I cannot personally assess the full methodological soundness of this peer-reviewed study, nor predict a realistic timeline for its eventual clinical application. My method is to faithfully report what the researchers themselves state in their publications and press materials, while clearly flagging the limitations and validation steps still needed before any change in clinical practice.

Sources

Primary sources

News-Medical — New breast cancer classification system predicts immunotherapy response success, July 3, 2026

EurekAlert! — Decoding tumor immunity: a four-step CIC-based subtyping strategy for precision immunotherapy, July 2, 2026

Secondary sources

Cancer Research Institute — 2026 Cancer Immunotherapy Insights + Impact Report, June 2026

PubMed — Immune Checkpoint Blockade and Emerging Combination Platforms in Breast Cancer: A Narrative Review, May 5, 2026

Frontiers in Immunology — Revisiting the standard of care for immune checkpoint inhibitors, April 17, 2026

Breast Cancer Research Foundation — ASCO 2026 Key Takeaways