Tagged: summary

Oncotarget

Artificial Intelligence Could Transform Nutrition Care for Cancer Patients

July 28, 2026

“We propose a multi-agent architecture for nutritional oncology governed by a graduated autonomy model, and discuss the evidentiary, regulatory, and equity barriers that must be addressed before clinical deployment.”

Nutrition plays a critical role in cancer care, yet it is often overlooked. Many patients experience weight loss, muscle wasting, treatment-related side effects, and changes in appetite that can affect their ability to tolerate therapy and maintain quality of life. Although nutrition specialists can help manage these challenges, access to specialized nutritional care remains limited in many healthcare settings.

An editorial published in Volume 17 of Oncotarget, titled “Artificial intelligence in nutritional oncology: From isolated screening tools to agentic intervention systems,” explores how advances in artificial intelligence (AI) could help address this gap. The editorial was written by Arnab Sarkar and corresponding author Yashbir Singh-Wolkenhauer, who is affiliated with the Department of Radiology, Mayo Clinic, Rochester, Minnesota. Rather than presenting new clinical trial data, the authors outline a future vision in which AI systems move beyond isolated tasks to continuously support nutritional care throughout a patient’s cancer journey. 

The Hidden Challenge of Cancer-Related Malnutrition

Cancer-related malnutrition is far more common than many people realize. Depending on the type and stage of cancer, it affects between 40% and 80% of patients and is estimated to contribute to 10% to 20% of cancer-related deaths. Cancer cachexia can develop in up to 80% of patients with advanced disease and is an independent predictor of mortality regardless of body mass index.

Professional organizations such as the European Society for Clinical Nutrition and Metabolism (ESPEN) and the American Society for Parenteral and Enteral Nutrition (ASPEN) recommend routine nutritional screening for patients with cancer. However, implementing these recommendations remains difficult. The authors note that only about one-quarter of oncology clinicians report having nutrition specialists integrated into their multidisciplinary teams, while the estimated ratio of registered dietitians to oncology patients in U.S. outpatient settings is approximately 1 to 2,308.

The authors argue that this is not simply a lack of nutritional knowledge, but a healthcare systems challenge that technology may help address.

AI Is Already Helping—but Only One Task at a Time

Artificial intelligence is already being applied to several aspects of nutritional oncology.

Machine learning models can identify patients at high risk for malnutrition with accuracy comparable to, or exceeding, traditional screening tools. Deep learning algorithms can analyze routine CT scans to automatically measure skeletal muscle and body composition, helping clinicians detect sarcopenia without requiring additional imaging. AI-powered virtual dietitians have also shown encouraging results, with patients reporting that they used the guidance to inform their diets and better manage treatment-related symptoms. However, these findings came from an observational deployment of a commercial product rather than a randomized trial and should be interpreted cautiously.

Despite these advances, the authors point out that today’s AI applications generally perform only one specific task at a time. One system may identify malnutrition risk, another may analyze body composition, while another provides dietary recommendations. These tools typically do not communicate with one another or continuously adapt as a patient’s condition changes during treatment.

From Individual Tools to Intelligent Care Partners

The editorial proposes moving beyond isolated AI applications toward agentic AI.

Unlike conventional AI models that respond to individual requests, agentic AI systems are designed to reason through complex problems, use multiple information sources, plan future actions, and remember previous interactions. Rather than answering a single question such as whether a patient is malnourished, an agentic system could continuously work toward the broader goal of optimizing a patient’s nutritional status throughout cancer treatment.

According to the authors, these systems could integrate information from electronic health records, laboratory results, medical imaging, dietary records, wearable devices, and clinical guidelines while adapting recommendations as treatments and symptoms evolve.

A Vision for AI-Assisted Nutritional Oncology

The authors describe a proposed multi-agent architecture in which several specialized AI agents work together under coordinated human oversight. The diagram presented in the editorial illustrates four specialized agents connected through a central coordination agent.

Within this framework:

-A Nutritional Screening Agent could monitor laboratory values, weight changes, and CT-derived body composition to detect early signs of malnutrition or cachexia.

-A Dietary Planning Agent could generate personalized meal plans that account for treatment side effects, cultural preferences, and individual dietary needs.

-A Treatment–Nutrition Interaction Agent could evaluate potential interactions between medications, nutritional supplements, and chemotherapy schedules.

-A Patient Engagement Agent could provide ongoing coaching through mobile applications or text messaging while tracking food intake and wearable health data.

-A central Coordination Agent would integrate information from these specialized systems, resolve conflicting recommendations, and alert clinicians whenever human review is needed.

Rather than replacing healthcare professionals, the proposed system is intended to support oncologists and dietitians by organizing complex information and providing timely recommendations.

Keeping Humans in Control

The authors emphasize that AI should not be allowed to make all clinical decisions independently.

Instead, they propose a graduated autonomy model, in which the level of AI independence depends on the clinical risk involved. Low-risk tasks, such as providing recipe suggestions or educational materials, could operate with relatively little supervision. Moderate-risk decisions, such as adjusting calorie targets or recommending referral to a dietitian, would require additional safeguards and clinician review. High-risk interventions, including decisions about enteral or parenteral nutrition, would always require explicit clinician authorization.

This framework aims to balance the efficiency of AI with the need for patient safety and clinical oversight.

Challenges That Must Be Addressed

Although the concept is promising, the authors stress that significant challenges remain before agentic AI can become part of routine cancer care.

Current AI systems are not perfect and can still generate inaccurate recommendations. No autonomous AI agent has received clearance from the U.S. Food and Drug Administration for independent clinical decision-making, and no randomized controlled trials have yet evaluated AI-driven nutritional interventions against major oncology outcomes such as survival, treatment completion, or maintenance of treatment dose intensity.

The editorial also highlights concerns related to patient privacy, clinical responsibility, and algorithmic bias. Because dietary habits vary across cultures, religions, and socioeconomic groups, AI systems must be validated in diverse populations to ensure equitable care and avoid reinforcing existing healthcare disparities.

Looking Ahead

The authors conclude that many of the technological building blocks needed for agentic AI in nutritional oncology already exist. They argue that nutritional care represents an especially promising area for future AI development because of the high prevalence of cancer-related malnutrition, shortages of specialized nutrition professionals, and the growing ability of AI systems to integrate complex clinical information.

However, they also emphasize that rigorous clinical validation remains essential before these systems can be widely adopted. Future research will need to determine whether AI-guided nutritional interventions improve meaningful patient outcomes while ensuring safety, fairness, and appropriate human oversight.

As artificial intelligence continues to evolve, the authors envision a future in which intelligent clinical support systems help oncologists and dietitians deliver more personalized, timely, and coordinated nutritional care—making nutrition a more integrated part of comprehensive cancer treatment rather than an afterthought.

Click here to read the full editorial published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].

Oncotarget

Rare Genetic Combination in Gastrointestinal Tumor Offers New Insights Into Precision Cancer Care

July 13, 2026

“This case suggests that oncogenic KIT signaling may remain the dominant driver of GIST behavior despite the presence of a germline SDHC mutation and highlights the importance of integrated molecular interpretation in GIST management.”


Gastrointestinal stromal tumors (GISTs) are the most common mesenchymal tumors of the digestive tract. Although many GISTs can be effectively treated with targeted therapies, their response to treatment depends largely on the specific genetic alterations driving tumor growth. Advances in molecular testing have transformed the management of these tumors, allowing clinicians to tailor treatment based on each patient’s genetic profile.

A case report published in Volume 17 of Oncotarget, titled “Small bowel GIST harboring concurrent KIT exon 9 duplication and SDHC mutation: A case report,” describes an exceptionally rare case involving a patient whose tumor carried two genetic alterations that are traditionally considered mutually exclusive.

Understanding the Genetics of GIST

Most GISTs develop because of activating mutations in either the KIT or PDGFRA genes. These mutations continuously stimulate growth signals within tumor cells, making them highly responsive to targeted drugs known as tyrosine kinase inhibitors, particularly imatinib.

A smaller group of GISTs lack these mutations and instead develop through abnormalities involving the succinate dehydrogenase (SDH) complex. These SDH-deficient tumors tend to occur in younger patients, most often arise in the stomach, and generally respond poorly to imatinib.

Because these two molecular pathways are thought to represent distinct mechanisms of tumor development, tumors containing both alterations are considered extremely uncommon. This case adds to growing evidence that rare exceptions to this long-standing assumption can occur.

An Unusual Genetic Profile

The patient was a 68-year-old man who sought medical attention after experiencing several weeks of worsening abdominal pain, bloating, and constipation. Imaging revealed a large mass arising from the small intestine.

A biopsy confirmed that the tumor was a spindle-cell GIST. Comprehensive next-generation sequencing identified two notable genetic findings:

  • A KIT exon 9 A502_Y503 duplication, a mutation known to drive many small bowel GISTs and predict responsiveness to higher-dose imatinib.
  • A germline SDHC p.R50C mutation, an inherited alteration associated with hereditary syndromes involving GISTs and certain neuroendocrine tumors.

Finding both mutations in the same patient was highly unusual because KIT-driven and SDH-deficient GISTs have traditionally been regarded as separate molecular subtypes. Genetic counseling subsequently confirmed that the SDHC mutation was inherited rather than acquired within the tumor itself.

Targeted Therapy Produced a Strong Response

Because the tumor carried a KIT exon 9 mutation, the multidisciplinary care team recommended treatment with imatinib before surgery to shrink the tumor and improve the chances of complete removal.

The patient initially received standard-dose imatinib, followed by dose escalation to 800 mg daily, consistent with current recommendations for KIT exon 9-mutant GISTs.

After three months of treatment, imaging demonstrated a clear partial response. The tumor decreased in size, and positron emission tomography (PET) scans showed a marked reduction in metabolic activity, indicating that the cancer had become substantially less active.

Following six months of targeted therapy, surgeons successfully removed the tumor. Pathologic examination revealed extensive treatment-related necrosis with negative surgical margins, and follow-up imaging five months after surgery showed no evidence of disease recurrence.

Why the SDHC Mutation Did Not Appear to Drive the Tumor

One of the most intriguing aspects of the case was that the inherited SDHC mutation did not produce the biological features typically seen in SDH-deficient GISTs.

Normally, tumors caused by SDH deficiency lose expression of SDHB, a protein commonly used by pathologists as a marker of dysfunction within the SDH complex. In this patient’s tumor, however, SDHB expression remained intact despite the presence of the inherited SDHC mutation.

The researchers suggest that this occurred because only one copy of the SDHC gene was altered. Development of true SDH-deficient GIST generally requires loss of function in both gene copies, whereas activating KIT mutations can drive tumor growth even when only one allele is affected.

Taken together, the preserved SDHB expression, the tumor’s location in the small intestine, and its robust response to high-dose imatinib all supported the conclusion that the KIT mutation—not the SDHC mutation—was the dominant driver of the cancer.

Why This Case Matters

As next-generation sequencing becomes increasingly common in cancer care, clinicians are identifying more tumors that harbor multiple potentially important genetic alterations.

This report highlights an important principle of precision oncology: not every detected mutation necessarily drives tumor behavior or determines treatment response. Instead, molecular findings must be interpreted alongside pathology, imaging, clinical presentation, and treatment outcomes.

The case also demonstrates the value of comprehensive genomic testing. Although the inherited SDHC mutation did not appear to influence treatment of this GIST, identifying it allowed the patient to receive genetic counseling because germline SDHC mutations have been associated with hereditary conditions that increase the risk of other tumors, including paragangliomas and pheochromocytomas.

Looking Ahead

The authors conclude that this rare case expands current understanding of GIST biology by showing that a tumor harboring both a KIT exon 9 duplication and a germline SDHC mutation can still behave like a classic KIT-driven GIST and respond well to high-dose imatinib. While additional cases will be needed to determine how often these uncommon genetic combinations occur, the findings underscore the importance of integrating molecular testing with clinical and pathological evaluation when selecting targeted therapies.

As precision medicine continues to evolve, studies such as this illustrate that understanding which genetic alteration truly drives a tumor may be just as important as identifying every mutation it carries.

Click here to read the full case report published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].

Oncotarget

Targeted Therapies Have Coincided With a Dramatic Decline in Multiple Myeloma Mortality in the United States

June 29, 2026

“Our findings highlight the real-world impact of targeted therapies on population-level outcomes and underscore the urgent need for care models that ensure accessibility, affordability, and long-term sustainability in the era of precision oncology.”

Multiple myeloma is the second most common blood cancer in the United States and has long been considered a difficult disease to treat. For decades, treatment options were limited, and survival remained poor. However, the therapeutic landscape has changed dramatically over the past several decades with the introduction of stem cell transplantation, targeted drugs, immunotherapies, and more recently, CAR T-cell therapy and bispecific antibodies.

A research paper titled “Targeted therapeutics and U.S. population-level mortality trends in multiple myeloma: A SEER-based analysis from 1975 to 2023” was published in Volume 17 of Oncotarget. In this study, the researchers examined how these major treatment advances have coincided with changes in multiple myeloma mortality across the United States over nearly five decades. The study was led by first and corresponding author Navkirat Kahlon from the Mass General Cancer Center at Wentworth-Douglass Hospital in Dover, New Hampshire. 

Looking Back at Nearly Fifty Years of Progress

To understand how multiple myeloma outcomes have evolved over time, the researchers analyzed mortality data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program between 1975 and 2023. Using Joinpoint regression analysis, they evaluated age-adjusted mortality rates and identified periods during which mortality trends changed significantly.

Rather than examining individual clinical trials, the study looked at population-level mortality patterns and compared them with the timing of major therapeutic advances. Although this type of analysis cannot prove that new treatments directly caused changes in mortality, it can reveal whether improvements in survival occurred alongside the introduction of new therapies.

From Limited Treatment Options to Targeted Therapies

The earliest years of the analysis reflected an era when treatment options were largely restricted to alkylating agents and corticosteroids. During this period, population-level mortality from multiple myeloma increased steadily.

The first sustained decline in mortality appeared during the mid-1990s, coinciding with the adoption of high-dose chemotherapy followed by autologous stem cell transplantation. Over the following decades, additional improvements paralleled the introduction of several new therapeutic classes, including: Proteasome inhibitors, Immunomodulatory drugs (IMiDs), Monoclonal antibodies, Selective inhibitors of nuclear export (SINEs), CAR T-cell therapies, Bispecific antibodies.

Each successive wave of therapeutic innovation expanded treatment options and improved disease control for patients with multiple myeloma.

Mortality Declined as Treatment Options Expanded

The analysis identified several distinct periods in U.S. mortality trends.

Between 1975 and 1994, mortality increased significantly.

From 1994 to 2002, mortality began to decline following the introduction of autologous stem cell transplantation.

Between 2002 and 2009, mortality declined more rapidly during the period when proteasome inhibitors and immunomodulatory drugs entered routine clinical practice.

After a brief plateau between 2009 and 2014, mortality resumed declining from 2014 through 2021, coinciding with expanded use of monoclonal antibodies, maintenance therapy, and combination treatment strategies.

The most striking finding occurred during 2021–2023, when investigators observed the steepest decline in mortality across the entire study period, with an annual percent change of −5.64%. This period overlapped with the introduction of BCMA-directed CAR T-cell therapies and bispecific antibodies into clinical practice.

A New Era of Precision Immunotherapy

One of the major themes emerging from the study is how treatment for multiple myeloma has shifted from broadly acting chemotherapy toward increasingly targeted therapies.

Modern treatments act against myeloma cells through multiple mechanisms. Some inhibit the cellular machinery that cancer cells rely on for survival, others stimulate the immune system to recognize malignant plasma cells, while newer cellular therapies genetically engineer a patient’s own immune cells to target proteins expressed on myeloma cells.

These advances have transformed multiple myeloma from a disease with limited therapeutic options into one that can often be managed through multiple sequential lines of treatment, extending survival for many patients. The authors suggest that the cumulative effect of these therapeutic innovations likely contributed to the long-term decline in population-level mortality observed in the study.

Progress Brings New Challenges

Although survival has improved substantially, the researchers emphasize that important challenges remain.

Multiple myeloma is still considered incurable for most patients, requiring long-term treatment that may continue for years. Many of the newest therapies are also associated with considerable financial costs, specialized administration, and potential long-term toxicities.

The authors note that access to these advanced treatments remains uneven, with geographic location, socioeconomic status, insurance coverage, and availability of specialized treatment centers influencing who can benefit from recent therapeutic advances. They argue that improving access and affordability will be essential if future survival gains are to be shared more equitably across patient populations.

Looking Ahead

The authors conclude that mortality from multiple myeloma has declined substantially over the past five decades, with population-level improvements occurring alongside successive waves of therapeutic innovation. Their findings suggest that advances such as stem cell transplantation, targeted therapies, monoclonal antibodies, CAR T-cell therapy, and bispecific antibodies have collectively reshaped the treatment landscape.

While additional research will be needed to determine the long-term impact of newer immunotherapies and to address challenges related to survivorship, treatment costs, and healthcare access, this study illustrates how sustained scientific innovation can translate into measurable improvements in outcomes at the population level. As new therapies continue to emerge, ongoing monitoring of real-world mortality trends will remain essential for understanding how advances in cancer treatment continue to shape long-term outcomes for patients with multiple myeloma.

Click here to read the full research paper published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].

Oncotarget

PDX1 May Link Metabolic Dysfunction to Prostate Cancer Progression

June 16, 2026

“Overall, our findings suggest that PDX1 plays a tumor-promoting role in human PCa cells by influencing expression of metabolites in insulin, inflammatory, and epithelial-mesenchymal transition (EMT) signaling pathways.”

Prostate cancer is the most commonly diagnosed cancer among men and remains a leading cause of cancer-related death worldwide. While age, family history, and genetics are well-established risk factors, researchers have increasingly focused on the role of metabolic health in prostate cancer progression. Obesity, insulin resistance, type 2 diabetes, and chronic inflammation have all been associated with more aggressive disease, but the molecular mechanisms connecting these conditions to prostate cancer remain incompletely understood.

A research paper titled “Epigenetic dysregulation and biological function of PDX1 in prostate cancer” was published in Volume 17 of Oncotarget. In this study, the researchers investigated whether a gene best known for regulating pancreatic function may also play an important role in prostate tumor biology. The study was led by first author Tayo A. Adeyika and corresponding author Bernard Kwabi-Addo from Howard University, Washington, DC. 

A Metabolic Gene Enters the Prostate Cancer Spotlight

PDX1 (Pancreatic and Duodenal Homeobox 1) is a transcription factor that plays a critical role in pancreatic development and insulin production. In pancreatic beta cells, PDX1 helps regulate glucose homeostasis and insulin gene expression. Altered PDX1 activity has previously been linked to metabolic disorders, including obesity and type 2 diabetes.

However, far less is known about its role in prostate cancer.

The researchers became interested in PDX1 after analyzing genome-wide DNA methylation data from prostate tissues. They discovered that the PDX1 gene was significantly hypermethylated in prostate cancer compared with normal prostate tissue. Surprisingly, despite this increase in DNA methylation, prostate tumors also exhibited substantially higher PDX1 protein expression. Tissue microarray analysis revealed approximately 2.3-fold higher PDX1 expression in cancer tissues than in matched normal prostate samples.

This finding highlights the complexity of epigenetic regulation in cancer. While DNA methylation is often associated with gene silencing, methylation occurring within gene-body regions can sometimes correlate with increased gene expression. The authors suggest that this mechanism may help explain the elevated PDX1 expression observed in prostate tumors.

Evidence That PDX1 Promotes Tumor Growth

To determine whether PDX1 actively contributes to prostate cancer progression, the investigators manipulated PDX1 expression in two prostate cancer cell lines: androgen-dependent LNCaP cells and androgen-independent PC-3 cells.

When PDX1 was overexpressed, cancer cells proliferated more rapidly and demonstrated enhanced migratory behavior. In contrast, suppressing PDX1 using shRNA significantly reduced cell growth and impaired cell migration and invasion. These gain-of-function and loss-of-function experiments consistently pointed toward a tumor-promoting role for PDX1.

The findings suggest that PDX1 is not simply associated with prostate cancer but may actively participate in biological processes that support tumor expansion and dissemination.

Connecting Insulin Signaling and Cancer

One of the most notable findings was the effect of PDX1 on genes involved in insulin-related signaling pathways.

The researchers found that PDX1 overexpression increased expression of the insulin receptor (INSR) and insulin-like growth factor 1 receptor (IGF1R), two signaling molecules that have long been implicated in cancer cell growth, metabolism, and survival. Conversely, reducing PDX1 expression decreased the activity of these pathways.

These observations are particularly interesting because elevated insulin and IGF signaling are common features of obesity and insulin resistance. The results raise the possibility that PDX1 may help translate metabolic abnormalities into biological signals that promote prostate cancer progression.

The study also identified an inverse relationship between PDX1 and ESR2, the gene encoding estrogen receptor beta (ERβ), which has frequently been associated with tumor-suppressive effects in the prostate.

Driving Cellular Changes Associated With Metastasis

The investigators also explored whether PDX1 influences epithelial-mesenchymal transition (EMT), a process that enables cancer cells to become more mobile and invasive.

EMT is characterized by changes in gene expression that help tumor cells detach from their original location, invade surrounding tissues, and eventually spread to distant organs. The researchers found that increased PDX1 expression was associated with elevated levels of several important EMT regulators, including SNAI1, TWIST1, and CDH2. Suppressing PDX1 produced the opposite effect.

These findings suggest that PDX1 may contribute to metastatic potential by activating pathways that promote cellular plasticity and migration.

High Glucose Amplifies PDX1 Activity

Because PDX1 is fundamentally involved in glucose regulation, the researchers examined how prostate cancer cells responded to different glucose concentrations.

They discovered that high-glucose conditions intensified several PDX1-associated effects, particularly those related to proliferation, insulin signaling, and EMT. Cells overexpressing PDX1 showed greater proliferation and stronger activation of genes involved in insulin signaling and EMT when exposed to elevated glucose levels. In contrast, PDX1 knockdown cells exhibited reduced responsiveness to glucose stimulation.

This observation may have important clinical implications. Hyperglycemia is common among individuals with diabetes and metabolic syndrome, conditions that have been linked to poorer prostate cancer outcomes. The findings suggest that elevated glucose levels may create an environment that amplifies PDX1-driven tumor-promoting pathways.

Why These Findings Matter

The study places PDX1 at the intersection of several biological processes that are increasingly recognized as important in prostate cancer, including insulin signaling, inflammation, metabolic regulation, and epithelial-mesenchymal transition.

Rather than acting through a single pathway, PDX1 appears to function as a broader regulatory factor that influences multiple signaling networks involved in tumor growth and progression. Its ability to integrate metabolic and oncogenic signals may help explain why obesity, diabetes, and prostate cancer often appear biologically interconnected.

Looking Ahead

The authors conclude that PDX1 functions as an important regulator of prostate cancer biology and that its effects become particularly pronounced under high-glucose conditions. Their findings support a model in which PDX1 contributes to tumor growth, migration, and metabolic adaptation through coordinated effects on insulin signaling, inflammatory pathways, and EMT-related genes.

Although additional studies will be needed to determine whether PDX1 can be used clinically as a biomarker or therapeutic target, this work provides new insight into how metabolic dysfunction may influence prostate cancer progression. As researchers continue to explore the relationship between cancer and metabolism, PDX1 may represent an important molecular bridge connecting metabolic dysfunction and prostate cancer progression. 

Click here to read the full research paper published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].

Oncotarget

Cancer Care Often Overlooked in Humanitarian Crises 

June 3, 2026

“Cancer is an escalating yet neglected health crisis among refugees, migrants, and populations affected by conflict.”

Cancer is increasingly recognized as a major global health challenge, yet for people living through war, displacement, and humanitarian crises, access to even basic oncology services can be difficult or impossible. While emergency responses typically focus on trauma care, infectious diseases, and immediate survival needs, cancer care remains largely absent from many humanitarian health programs. 

A review paper on this topic was published in Volume 17 of Oncotarget titled “Cancer without borders: Policy frameworks for oncology care in humanitarian and conflict settings.” The study was led by first and corresponding author Pragnesh Parmar, with Gunvanti Rathod as co-author, both from AIIMS Bibinagar, Telangana, India.

Cancer Care in Humanitarian Emergencies 

The authors reviewed published research, humanitarian agency reports, and case studies from conflict-affected regions, including Gaza, Sudan, and Ukraine. Their analysis highlights how disrupted healthcare infrastructure, shortages of medical personnel, legal barriers, and fragmented healthcare systems can prevent patients from receiving timely cancer diagnosis and treatment.

According to the review, oncology services are frequently overlooked during humanitarian emergencies because health systems are often forced to prioritize immediate life-threatening conditions. However, the authors note that non-communicable diseases, including cancer, account for a substantial portion of the disease burden among displaced populations.

Barriers to Diagnosis and Treatment 

The review identifies several challenges that cancer patients commonly face in humanitarian settings. These include interruptions in diagnostic services, limited access to chemotherapy and radiotherapy, lack of specialist care, financial hardship, and difficulties crossing borders to receive treatment.

The authors also emphasize that vulnerable groups—including women, children, and older adults—often face additional obstacles to obtaining care. In many settings, social stigma, transportation difficulties, and legal restrictions can further delay diagnosis and treatment.

Lessons From Conflict Zones 

The paper examines experiences from several regions affected by conflict. In Gaza, restrictions on movement and limited local oncology resources have complicated access to specialized cancer treatment. In Sudan, ongoing conflict has severely disrupted cancer services and damaged healthcare infrastructure. In contrast, the response to displaced cancer patients from Ukraine demonstrated how international coordination and cross-border referral systems can help maintain continuity of care during crises.

These examples illustrate how healthcare outcomes can vary dramatically depending on available infrastructure, policy support, and international collaboration.

The Role of Technology 

The review also discusses emerging approaches that may improve access to oncology care in crisis settings. Tele-oncology programs allow healthcare providers to consult with cancer specialists remotely, while mobile medical units can bring screening and diagnostic services closer to displaced populations.

According to the authors, these strategies may help bridge gaps in access where traditional oncology facilities are unavailable or inaccessible.

Policy Changes Needed 

The authors argue that cancer should no longer be viewed as a secondary concern during humanitarian emergencies. They recommend integrating oncology services into emergency response planning, developing cross-border treatment agreements, expanding telemedicine programs, and strengthening international cooperation to support displaced cancer patients.

They also highlight the need for better cancer surveillance systems and more comprehensive data collection to understand the true burden of cancer among refugees and conflict-affected populations.

Looking Ahead 

The authors conclude that improving cancer care in humanitarian settings will require coordinated action from governments, international organizations, healthcare providers, and humanitarian agencies. Expanding access to diagnosis, treatment, and palliative care could help reduce preventable suffering among millions of displaced people worldwide.

As the authors write, “Addressing cancer in humanitarian contexts is not merely a technical challenge but a moral imperative.” Their findings suggest that ensuring access to cancer care should become a central component of global humanitarian health efforts rather than an afterthought.

Click here to read the full review published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].

Oncotarget

Immune Checkpoint Inhibitors May Improve Outcomes in High-Risk Solid Tumors

May 20, 2026

“This systematic review and meta-analysis of 13 randomized controlled trials including 9,850 patients shows that adjuvant PD-1 and PD-L1 inhibitors improve disease-free and distant metastasis-free survival in patients with high-risk solid tumors.”

Cancer immunotherapy has transformed the treatment landscape for many advanced cancers over the past decade. Drugs targeting the PD-1 and PD-L1 pathways are now widely used across several tumor types, helping the immune system recognize and attack cancer cells more effectively. However, researchers are still working to understand how beneficial these therapies may be when used earlier in the disease course, particularly after surgery in patients with high-risk solid tumors.

A research paper on this topic was published in Volume 17 of Oncotarget titled “Efficacy and safety of PD-1/ PD-L1 inhibitors as adjuvants in the treatment of patients with solid cancers: A systematic review and meta-analysis of randomized controlled trials.” 

Understanding Adjuvant Immunotherapy

Adjuvant therapy refers to treatment given after primary treatment, such as surgery, to help reduce the risk of cancer recurrence. While PD-1 and PD-L1 inhibitors are already established therapies for several advanced cancers, their role in earlier-stage solid tumors remains an active area of investigation.

To better evaluate their effectiveness and safety, the researchers conducted a systematic review and meta-analysis of 13 randomized controlled trials published between 2021 and 2023. Altogether, the analysis included 9,850 patients with cancers such as renal cell carcinoma, melanoma, non-small cell lung cancer, esophageal cancer, gastroesophageal junction cancer, and urothelial carcinoma.

Improved Disease-Free Survival

The analysis found that adjuvant PD-1 and PD-L1 inhibitors were associated with improved disease-free survival and distant metastasis-free survival compared with control groups. In several studies, therapies such as pembrolizumab and nivolumab reduced the likelihood of cancer recurrence or distant metastasis in patients with high-risk tumors.

The authors also noted that treatment responses varied depending on cancer type and therapeutic regimen, highlighting the complexity of immune responses across different tumor environments.

Balancing Benefits and Side Effects

Although the therapies improved several important clinical outcomes, they were also associated with increased adverse events. Commonly reported side effects included fatigue, diarrhea, rash, nausea, pruritus, hypothyroidism, and arthralgia.

The study found that treatment-related adverse events occurred more frequently in patients receiving PD-1 and PD-L1 inhibitors compared with control groups. The authors emphasized that careful monitoring and long-term follow-up remain important when using these therapies in the adjuvant setting.

Why Longer Follow-Up Matters

One important finding from the analysis was that, despite improvements in disease-free survival, a clear overall survival benefit has not yet been consistently demonstrated. According to the authors, this may partly reflect the relatively short follow-up periods currently available in many of the trials.

The researchers also noted that the number of studies within each cancer subtype remains limited, making cancer-specific conclusions more difficult at this stage.

Looking Ahead

This study highlights the growing potential of adjuvant immunotherapy in solid cancers while also underscoring the need for longer-term data and more precise patient selection strategies. Future studies will likely focus on identifying which patients benefit the most, refining biomarker-guided treatment approaches, and improving management of immune-related side effects.

Still, the findings suggest that PD-1 and PD-L1 inhibitors may play an increasingly important role in reducing recurrence risk in selected patients with high-risk solid tumors.

The authors concluded that larger trials with longer follow-up are needed to better define which patient groups derive the greatest benefit.

Click here to read the full research paper published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].

Oncotarget

Mapping the Hidden Structure of Glioma Research: What Are We Missing?

May 5, 2026

“Unlike previous studies that focused primarily on metrics such as the h-index, our approach identifies the limited but notable mention of social factors in glioma classification research, thereby highlighting a thematic gap.”

Glioma research has evolved rapidly over the past decade, driven by breakthroughs in molecular biology, imaging technologies, and computational tools. Today, clinicians can classify tumors with far greater precision than ever before, using genetic mutations, epigenetic markers, and advanced diagnostic frameworks. Yet, despite this progress, an important question remains: are we truly capturing the full picture of what shapes patient outcomes?

Traditionally, glioma classification has focused on what can be measured in the tumor itself—its histology, molecular profile, and biological behavior. While these factors are undeniably critical, they may not fully explain why patients with similar tumors can experience very different clinical trajectories. Increasingly, researchers are beginning to recognize that broader influences—particularly social and environmental factors—may also play a role. Understanding how these different layers of information connect is becoming an important challenge in neuro-oncology.

A review was published in Volume 17 of Oncotarget on March 31, 2026, titled “Bibliometric mapping of glioma classification research through main path, key route, and K-core analyses.” The study was led by first and corresponding author Kayode Ahmed from The University of Texas MD Anderson Cancer Center, in collaboration with Juan E. Núñez-Ríos from Universidad Panamericana. 

A Bird’s-Eye View of an Entire Field

Rather than focusing on a single experiment or dataset, the researchers analyzed the structure of glioma research itself. Using bibliometric and network-based approaches, they examined thousands of scientific publications to understand how knowledge in this field has developed over time.

By constructing a large citation network—comprising tens of thousands of articles and hundreds of thousands of connections—they were able to trace the intellectual pathways that have shaped modern glioma classification. Techniques such as main path analysis and key route analysis helped identify the most influential studies, while K-core analysis revealed tightly connected clusters of research activity.

What Drives Glioma Classification Today?

The findings confirm what many in the field might expect: glioma classification has been heavily shaped by advances in molecular and technological approaches.

DNA methylation profiling, genetic mutations, and imaging innovations have emerged as central pillars in modern classification systems. These tools have significantly improved diagnostic accuracy and helped refine prognostic models, enabling more personalized approaches to treatment.

At the same time, the network analysis highlights how interconnected these advances are. Progress in glioma research has not occurred in isolation, but through the convergence of multiple disciplines—molecular biology, bioinformatics, imaging science, and clinical oncology.

The Missing Piece: Social Factors

However, one of the most striking insights from the study is not what dominates the field—but what is largely absent.

Despite growing awareness that factors such as socioeconomic status, education, and access to healthcare can influence disease outcomes, these variables are rarely integrated into glioma classification research. Compared to molecular and imaging-based studies, the contribution of social determinants remains minimal.

This gap suggests that current classification systems, while biologically sophisticated, may still overlook important aspects of patient reality. As a result, they may fall short of fully explaining differences in treatment response and survival.

Why This Matters

This study shifts the conversation from what we know to how we know it. By mapping the structure of scientific research itself, the authors reveal both the strengths and blind spots of the field.

Glioma classification has become increasingly precise at the molecular level—but precision medicine may ultimately require a broader perspective. Integrating biological data with social and environmental context could lead to more comprehensive and clinically meaningful classification systems.

Looking Ahead

Moving forward, the challenge will be to bridge this gap. Incorporating social determinants into glioma research will not be straightforward, but it may be essential for developing more holistic models of disease.

Future studies may need to combine traditional biomedical approaches with data from public health, epidemiology, and social sciences. Such integration could improve not only how gliomas are classified, but also how patients are treated and supported.

Conclusion

By stepping back and examining the structure of glioma research as a whole, this study offers a fresh perspective on a rapidly evolving field. It highlights how far we have come in understanding the biology of gliomas—while also reminding us that important pieces of the puzzle may still be missing.

In the end, advancing glioma classification may depend not just on better technology, but on a more complete understanding of the patient behind the tumor.

Click here to read the full review published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].

Oncotarget

SCD1 Inhibition Strategy Shows Potent Synergy with Regorafenib and Metformin in Tumor Cell Killing

April 21, 2026

“Our data demonstrates that UM cells are killed by treatment with aramchol plus regorafenib plus metformin via enhanced autophagic flux and that this combination may have the potential to control UM tumors that have metastasized to the liver.”

Cancer has long been understood through a variety of biological frameworks, including genetic mutations, dysregulated signaling pathways, and uncontrolled cell proliferation. Yet, these models often capture the visible consequences of disease rather than the deeper metabolic dependencies that sustain tumor survival. Despite major advances in targeted therapies, a central challenge remains: what underlying mechanisms make cancer cells vulnerable to treatment, and how can these vulnerabilities be exploited more effectively? Increasing attention has shifted toward cellular metabolism—particularly lipid regulation and energy-sensing pathways such as AMPK—as critical determinants of tumor behavior. Scientists are now taking a closer look at how metabolism works together with stress responses like autophagy—and how this connection could be used to develop better cancer treatments.

A new research paper was published in Volume 17 of Oncotarget, titled “The SCD1 inhibitor aramchol interacts with regorafenib and metformin to kill tumor cells.” The study was led by first author Michael R. Booth and corresponding author Paul Dent from Virginia Commonwealth University, in collaboration with Laurence Booth and Jane L. Roberts from Virginia Commonwealth University and John M. Kirkwood from the University of Pittsburgh Cancer Institute.

Targeting Tumor Metabolism Through Lipid Enzymes

Cancer cells rely heavily on metabolic reprogramming to sustain growth and survival. One enzyme of growing interest is stearoyl-CoA desaturase 1 (SCD1), which regulates lipid metabolism and cellular redox balance. Aramchol, an SCD1 inhibitor originally developed for metabolic liver disease, is being investigated as a potential strategy for targeting tumor metabolism.

In this study, researchers explored how aramchol behaves in combination with two clinically relevant agents: regorafenib, a multi-kinase inhibitor, and metformin, a widely used anti-diabetic drug known to activate AMPK signaling.

A Three-Drug Combination with Enhanced Anti-Tumor Activity

The findings reveal a clear hierarchy of therapeutic impact. While aramchol alone showed modest anti-tumor activity, its combination with regorafenib significantly increased tumor cell death. Notably, the addition of metformin further amplified this effect, producing the strongest response across multiple tumor models, including uveal melanoma (UM) and cholangiocarcinoma cells.

This enhanced killing effect was associated with coordinated signaling changes, including AMPK activation and suppression of mTOR-related pathways—key regulators of cellular energy balance and survival.

Autophagy: A Central Mechanism of Tumor Cell Death

A defining feature of the study is the identification of autophagic flux as a central mechanism underlying tumor cell killing.

Using LC3-based fluorescence assays, the authors demonstrated that the drug combination markedly increased the formation of autophagosomes and autolysosomes, indicating robust activation of macroautophagy. Importantly, silencing essential autophagy regulators such as Beclin1, ATG5, and LAMP2 significantly reduced both autophagic activity and tumor cell death.

These findings suggest that autophagy is an important functional component of therapeutic efficacy in this context.

Dual Mechanisms: Autophagy and Death Receptor Signaling

Beyond autophagy, the study also highlights a second critical pathway: death receptor signaling via BID.

The researchers showed that BID knockdown reduced tumor cell killing, supporting a role for BID-dependent death signaling alongside macroautophagy in the response to treatment. Together, these findings suggest that the drug combination engages multiple stress-response pathways that contribute to tumor cell death.

Re-Evaluating the Role of SCD1

Although aramchol targets SCD1, the study provides an important nuance: SCD1 inhibition alone does not fully explain the observed anti-cancer effects.

While SCD1 knockdown modestly increased baseline cell death, it did not replicate the full potency of the drug combination. This suggests that aramchol likely engages additional molecular targets or pathways, expanding its therapeutic relevance beyond lipid metabolism alone.

Implications for Uveal Melanoma and Liver Metastasis

Uveal melanoma is a rare but aggressive cancer with a strong tendency to metastasize to the liver—a site where treatment options remain limited. Because aramchol concentrates in the liver, the findings suggest that this combination may hold particular relevance for metastatic UM, which most often spreads to that organ.

Looking Ahead

This study provides preclinical evidence that combining aramchol, regorafenib, and metformin can significantly enhance tumor cell killing through coordinated metabolic and stress-response pathways.

Future work will be needed to validate these findings in vivo and determine their clinical applicability. However, the results already point toward a promising strategy: integrating metabolic targeting with established anti-cancer therapies to overcome resistance and improve efficacy.

Conclusion

By uncovering how SCD1 inhibition synergizes with kinase inhibition and metabolic modulation, this study advances a more integrated view of cancer therapy. Rather than relying on single-target approaches, the findings emphasize the power of multi-pathway disruption—combining autophagy, apoptosis, and metabolic stress—to drive tumor cell death.

Click here to read the full research paper published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].

Oncotarget

CREB5: A Master Regulator of Stem Cell-Like Programs in Prostate Cancer Progression

April 7, 2026

“In this study, through both computational and molecular characterization of PC cell lines, we determined that CREB5 is associated with basal PC and drives SCL traits.”

Androgen receptor (AR) signaling has long been the central driver of prostate cancer progression and the primary target of therapies for advanced disease. Yet, a significant subset of tumors either fail to respond or develop resistance, often by switching to AR-independent programs that resemble basal or stem cell-like states. Understanding what drives these aggressive, therapy-resistant phenotypes is a critical challenge in oncology.

A research paper, titled “CREB5 regulates stem cell-like transcriptional programs to enhance tumor progression in prostate cancer” was published in Volume 17 of Oncotarget by a multi-institutional team of researchers, identifies a key molecular regulator of this process and reveals how it promotes tumor progression.

The work was led by first author Allison Makovec from the Department of Medicine and the Masonic Cancer Center at the University of Minnesota – Twin Cities, and University of Kansas Medical Center, along with corresponding authors Emmanuel S. Antonarakis and Justin Hwang (both from the Department of Medicine and the Masonic Cancer Center at the University of Minnesota – Twin Cities). The team’s investigation demonstrates that the transcription factor CREB5 drives basal and stem cell-like transcriptional programs, interacts with AP-1 proteins, and enhances tumor-forming capacity in prostate cancer cells.

The Discovery: CREB5 Links to Basal and Stem Cell-Like Programs

The researchers began by analyzing transcriptomic data from 493 primary prostate tumors (TCGA) and 208 castration-resistant prostate cancers (CRPC) from the SU2C dataset. They ranked approximately 20,000 genes based on their correlation with gene signatures defining luminal, basal, club, and hillock epithelial cell identities.

CREB5 ranked among the top genes associated with basal, club, and hillock identities but was among the lowest associated with luminal identity—the opposite pattern of AR itself. In both primary and CRPC samples, CREB5 expression was inversely correlated with AR activity and positively correlated with KLF5, a transcription factor previously linked to AR-independent resistance.

Further analysis revealed that CREB5-high tumors had significantly lower expression of AR, FOLH1 (PSMA), KLK2, and KLK3 (PSA) compared to CREB5-low tumors. Interestingly, AR-V7—a constitutively active AR splice variant that drives therapy resistance—was also decreased in CREB5-high tumors, suggesting that CREB5 operates through AR-independent pathways rather than AR splice variants.

Molecular Associations: CREB5 and the AP-1 Network

To understand how CREB5 drives these transcriptional programs, the team used the Algorithm for Linking Activity Networks (ALAN), which compares gene behavior across all potential interactions. CREB5 showed highly concordant behavior with the 25 transcription factors that define the stem cell-like (SCL) subtype of CRPC, as previously defined by Tang et al. Notably, CREB5 exhibited nearly identical behavior to FOSL1, a key AP-1 transcription factor implicated in stem cell features, therapy resistance, and metastasis in other cancers.

This relationship was remarkably strong. In CRPC samples, CREB5 and FOSL1 expression were significantly correlated (r = 0.47, p < 0.001), and ALAN analysis showed an R² of 0.980 between their gene behaviors across all genes in the dataset. Even in benign prostate tissue (GTEx), the alignment remained strong (R² = 0.707).

Functional validation confirmed the regulatory relationship. In LNCaP cells overexpressing CREB5, RNA-seq showed increased FOSL1 expression across multiple conditions—including androgen deprivation (CSS), enzalutamide treatment, and androgen stimulation (R1881). In CWR-R1 cells, CREB5 overexpression significantly increased FOSL1 expression by RT-qPCR (p = 0.014).

Mechanisms: CREB5 Interacts with AP-1 Proteins and Binds Their Regulatory Elements

To determine how CREB5 exerts its effects, the team examined protein-protein interactions using rapid immunoprecipitation and mass spectrometry of endogenous proteins (RIME) from prior work. Compared to controls, CREB5 interacted with several AP-1 factors, including JUN, JUNB, JUND, ATF2, and ATF7.

Motif enrichment analysis of CREB5 binding sites (from ChIP-sequencing) revealed significant enrichment of AP-1 binding motifs, including those for JUN, JUNB, and ATF2. In enzalutamide-treated cells, CREB5-bound sites remained enriched near ATF2 motifs, and CREB5 binding patterns were highly consistent at these sites.

ChIP-sequencing further showed that CREB5 bound to the transcriptional start sites of several AP-1 genes, including ATF3 and FOSL2, and to FOSL1 itself—particularly in enzalutamide-treated cells. Moreover, CREB5 bound to the transcriptional start and end sites of all 25 SCL genes defined by Tang et al., confirming its role as a broad regulator of stem cell-like transcriptional programs.

Phenotypic Consequences: CREB5 Drives Tumor-Forming Capacity

If CREB5 promotes stem cell-like traits, it should enhance the ability of cancer cells to form tumors. The team tested this using 3D tumorsphere assays in three cell lines: LNCaP (AR-positive, hormone-sensitive), CWR-R1 (CRPC-like), and CWR-R1 enzalutamide-resistant (enzR).

CREB5 overexpression significantly increased the number of tumorspheres in LNCaP cells compared to luciferase (LUC) controls (p < 0.01), indicating enhanced tumor-forming capacity from single cells. This effect was not observed in the more aggressive CWR-R1 or CWR-R1 enzR lines, suggesting that CREB5 has the greatest impact in hormone-sensitive cells, consistent with prior studies.

In vivo, LNCaP cells with CREB5 overexpression were implanted into castrated and non-castrated male mice. After 56 days, CREB5 overexpression significantly increased tumor volume in both castrated (p = 0.002) and non-castrated (p = 0.008) mice. Notably, there was no significant effect on tumor growth rate, supporting the hypothesis that CREB5 promotes tumor formation (stemness) rather than simply accelerating proliferation.

Clinical Implications and Future Directions

These findings have several important implications. First, they identify CREB5 as a central regulator of lineage plasticity in prostate cancer—the ability of tumor cells to switch from an AR-driven luminal identity to an AR-independent basal or stem cell-like state. This plasticity is a major mechanism of resistance to AR-targeted therapies.

Second, the inverse relationship between CREB5 and AR activity was detectable even in primary, treatment-naïve tumors. This suggests that high CREB5 expression may serve as a future biomarker for identifying patients at risk of developing resistance or progressing to metastatic disease, even before therapy begins.

Third, the interaction between CREB5 and AP-1 transcription factors—particularly FOSL1—points to potential therapeutic strategies. AP-1 factors are known regulators of cancer cell plasticity across multiple malignancies, and there are now anti-cancer therapies targeting AP-1 factors. Whether such agents can perturb CREB5’s tumor-promoting activity remains an open question.

The authors acknowledge that the mechanistic relationship between CREB5 and KLF5—another SCL-associated transcription factor—remains unclear, as no direct biochemical interaction was detected. Future studies will need to explore whether these factors operate in parallel pathways or through indirect mechanisms.

Future Perspectives and Conclusion

This study does not claim to have fully mapped the regulatory network of CREB5 in prostate cancer. Rather, it establishes CREB5 as a key driver of basal and stem cell-like transcriptional programs and provides a mechanistic link to AP-1 transcription factors.

The perspective that emerges is one where lineage plasticity in prostate cancer is not a random event but is driven by specific transcriptional regulators like CREB5. By integrating computational modeling, molecular biology, and functional studies, the team demonstrates that CREB5 enhances tumor-forming capacity through interactions with AP-1 factors and regulation of SCL genes.

Continued research will be needed to determine whether targeting CREB5 or its interaction with AP-1 complexes can mitigate deadly stem cell-like phenotypes in prostate cancer and potentially in other malignancies where CREB5 has been implicated—including breast, colorectal, ovarian, and brain cancers. As the authors note, “increased CREB5 may lead to specific mechanisms of therapy response, and future therapeutic strategies may consider antagonizing CREB5 interactions with AP-1 complexes.”

Click here to read the full research paper published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].

Oncotarget

Predicting Colorectal Cancer Survival: How Machine Learning Combines Clinical and Biological Clues

March 25, 2026

“Understanding both the biological and clinical aspects of the patient is essential to uncover the mechanism underlying the prognosis of the disease.”

Colorectal cancer (CRC) ranks among the most common and lethal cancers worldwide, accounting for approximately 10% of all cancer diagnoses. While advances in prevention and treatment have improved outcomes, predicting which patients will survive remains a complex challenge—one that depends on an intricate interplay between molecular biology and clinical factors.

A research paper, titled “Machine learning-based survival prediction in colorectal cancer combining clinical and biological features” was published in Volume 16 of Oncotarget by an international team of researchers, demonstrating how machine learning can integrate these two domains to achieve highly accurate survival predictions.

The team’s investigation demonstrates that combining clinical features—such as pathological stage, age, and lymph node status—with biological markers—including the E2F8 gene and hsa-miR-495-3p—can significantly improve the ability to predict patient survival.

The Method: Integrating Clinical and Biological Data

The researchers constructed a three-phase pipeline using data from 545 colorectal cancer patients from The Cancer Genome Atlas (TCGA) database. The data spanned colon, rectum, and rectosigmoid junction cancers, with patient ages ranging from 31 to 90 years.

In the first phase, data pre-processing, the team extracted and normalized both clinical and biological features. For biological features, they performed differential expression analysis, constructed competing endogenous RNA (ceRNA) networks, and conducted survival analysis to identify 19 candidate molecules—including mRNAs, lncRNAs, and miRNAs—with potential roles in CRC prognosis. For clinical features, they selected 13 characteristics, including age, pathological stage, lymph node counts, chemotherapy status, and new tumor events.

To handle missing data, they created three distinct cases: Case 1 filtered out missing biological or core clinical features; Case 2 also excluded patients with missing demographic features like race and weight; and Case 3 replaced missing values with the most frequent category.

In the second phase, feature selection, the team applied LASSO (Least Absolute Shrinkage and Selection Operator) to rank features by importance, followed by SHAP (Shapley Additive Explanations) to understand each feature’s impact on survival prediction.

In the third phase, model construction, they trained and compared six machine learning classifiers: Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), AdaBoost (AB), Stacking, and Voting.

Key Findings: Features That Matter Most

Across the three data cases, certain features consistently emerged as critical for predicting survival.

Among biological features, E2F8 stood out as the most significant, appearing in all three models. This gene, known to be associated with cell proliferation and CRC staging, has been identified by other studies as a potential CRC biomarker. WDR77 and hsa-miR-495-3p also proved important in most groups, consistent with previous research linking them to cancer development.

Among clinical features, pathological stage consistently ranked as the most influential predictor. Higher stage correlated strongly with lower survival probability. Age, new tumor event (likely representing recurrence), lymph node count, and chemotherapy status also emerged as critical factors.

Notably, the study identified that the combination of these features outperformed models relying on clinical or biological data alone.

Predictive Performance: Accuracy Reaches 89.58%

The machine learning models achieved impressive results. For Case 1 (filtered for core clinical features), an SVM model achieved 86.87% accuracy with an AUC of 83.49%. For Case 2 (more strictly filtered), an AdaBoost model achieved the best overall performance: 89.58% accuracy, though with a lower AUC of 76.50% due to dataset size limitations. For Case 3 (with imputed missing values), a Voting ensemble achieved 82.57% accuracy.

Bootstrap analysis confirmed that these advanced models provided meaningful improvements over baseline logistic regression, with accuracy increases ranging from 4.6% to 11.1%.

Biological Insights: The ceRNA Network Perspective

The 19 candidate molecules used as biological features were not chosen arbitrarily. They originated from a prior analysis by the same research group that constructed competing endogenous RNA (ceRNA) networks—complex regulatory systems where mRNAs, lncRNAs, and miRNAs cross-regulate each other through shared microRNA response elements.

This ceRNA framework is particularly relevant in cancer, where disruptions to these networks can drive tumor progression. By incorporating molecules from these networks, the study captured not just individual biomarkers but the broader regulatory context in which they operate.

Clinical Implications and Future Directions

The study’s findings carry several implications for clinical practice and future research.

First, they validate the prognostic value of well-established clinical factors—age, stage, lymph node status—while also highlighting novel molecular markers like E2F8 that warrant further investigation. Second, they demonstrate that machine learning can effectively integrate diverse data types to generate clinically useful predictions. Third, they underscore the importance of complete data collection; missing clinical information, such as race and weight, limited the analysis and may introduce bias.

The authors acknowledge limitations, including the relatively small dataset (545 patients), the exclusive use of US-based TCGA data, and the lack of experimental validation for the identified biomarkers. They call for future studies with larger, more diverse cohorts and for further investigation into the molecular mechanisms linking E2F8, miR-495-3p, and WDR77 to CRC prognosis.

Future Perspectives and Conclusion

This study does not claim to have developed a clinically deployable tool. Rather, it offers a proof-of-concept that machine learning can meaningfully integrate clinical and biological data to predict colorectal cancer survival. By combining LASSO feature selection with SHAP interpretability and ensemble modeling, the team demonstrates a pipeline that balances predictive power with biological insight.

The perspective that emerges is one where the future of cancer prognosis lies not in choosing between clinical or molecular data, but in systematically combining them. As the authors note, even basic patient information—age, weight, lymph node status—when accurately recorded and integrated with molecular profiles, can contribute powerfully to our understanding of disease trajectory.

Continued research will be needed to validate these findings in independent cohorts, to expand the set of biological features, and ultimately to translate these models into tools that can guide treatment decisions and improve outcomes for patients with colorectal cancer.

Click here to read the full research paper published in Oncotarget.

_______

Oncotarget is an open-access, peer-reviewed journal that has published primarily oncology-focused research papers since 2010. These papers are available to readers (at no cost and free of subscription barriers) in a continuous publishing format at Oncotarget.com. 

Oncotarget is indexed and archived by PubMed/Medline, PubMed Central, Scopus, EMBASE, META (Chan Zuckerberg Initiative) (2018-2022), and Dimensions (Digital Science).

Click here to subscribe to Oncotarget publication updates.

For media inquiries, please contact [email protected].