The clinical foundation of TIL therapy rests on a deceptively simple concept. Tumor-infiltrating lymphocytes, primarily cytotoxic CD8-positive T cells and helper CD4-positive T cells, migrate from peripheral blood into the tumor microenvironment, where they recognize and attack malignant cells through direct cytotoxicity. Adoptive cell therapy with autologous TILs amplifies this natural response: tumor-resident T cells are isolated from excised tumor tissue, expanded to billions in the laboratory, and returned to the patient after a non-myeloablative lymphodepleting chemotherapy regimen designed to clear immunosuppressive cells and cytokine sinks. The protocol that underpins most current regimens, including the FDA-approved product lifileucel, traces directly to the pioneering work of Steven Rosenberg’s laboratory at the U.S. National Cancer Institute, with only minor modifications accumulated over decades.

The clinical evidence base is now substantial. In 2024, the U.S. Food and Drug Administration approved lifileucel as the first autologous TIL product for patients with advanced cutaneous melanoma, a landmark moment for the field. A meta-analysis of 13 studies employing high-dose interleukin-2 in TIL therapy, encompassing 617 patients with advanced cutaneous melanoma including some with brain metastases, reported an objective response rate of 34 percent among patients previously treated with anti-PD-(L)1 agents and 44 percent among those without such prior treatment. Notably, the analysis found no statistically significant difference in objective response or complete response rates between the two groups, suggesting that interleukin-2’s immunomodulatory effects allow TIL therapy to retain benefit even after checkpoint inhibitor failure. Recent studies of patients refractory to immune checkpoint inhibitors or targeted therapies have shown objective response rates of 28 to 49 percent in melanoma resistant to standard treatments.

The reach of TIL therapy now extends well beyond melanoma. In non-small cell lung cancer, the IOV-COM-202 trial tested lifileucel in two cohorts following the same nonmyeloablative lymphodepletion regimen of cyclophosphamide and fludarabine, a single TIL infusion, and up to six doses of high-dose interleukin-2. Cohort 3B, comprising 28 patients with immunosuppressant-resistant metastatic disease, achieved an objective response rate of 21.4 percent after a median follow-up of 16 months. Cohort 3A, in checkpoint inhibitor-naive patients receiving pembrolizumab before lymphodepletion, showed a striking 64.3 percent objective response rate in the EGFR-wildtype subgroup at a median follow-up of 25.6 months. In cervical cancer, a National Cancer Institute phase 2 trial of 18 patients with HPV16- or HPV18-positive disease reported a 28 percent response rate, including two durable complete responses lasting 67 and 53 months, while a multicenter study by Iovance Biotherapeutics in checkpoint-naive advanced cervical cancer achieved a 44 percent objective response rate among 27 patients. Meta-analytic estimates suggest objective response rates of up to 26.3 percent across gynecologic cancers, and early signals of activity have emerged in colorectal cancer, cholangiocarcinoma, breast cancer, head and neck squamous cell carcinoma, and pancreatic ductal adenocarcinoma.

Understanding why interleukin-2 is both indispensable and hazardous requires a look at its receptor biology. The interleukin-2 receptor is a heterotrimeric protein composed of alpha (CD25), beta (CD122) and gamma (CD132) subunits, whose variable expression across immune cell types produces the cytokine’s diverse functional effects. Preclinical work has established the beneficial face of interleukin-2 signaling. Trentin and colleagues, studying TILs from 16 patients with solid tumors, demonstrated that interleukin-2 signaling through CD122 and CD25 promotes TIL proliferation and functional activation, marked by increased activity in the JAK/STAT pathway. Interleukin-2 also enhances cytotoxic function by boosting secretion of interferon-gamma and tumor necrosis factor-alpha, cytokines that directly inhibit tumor growth. In one in vitro study, interleukin-2 exposure led to downregulation of PD-1 on CD8-positive tumor-specific TILs, hinting at a mechanism for potentiating anti-tumor activity. The historical arc began in 1986, when Rosenberg’s group showed in murine models that interleukin-2 supplementation improves TIL efficacy, and culminated in a 1988 clinical trial combining TIL therapy with high-dose interleukin-2 that achieved a 60 percent objective response rate in previously untreated advanced melanoma patients. The standard regimen that followed, delivering at least 720,000 international units per kilogram intravenously every eight hours until dose-limiting toxicity or tolerance, has anchored metastatic melanoma treatment for more than 30 years.

The adverse face of interleukin-2, however, is impossible to ignore. In a clinical study of 391 patients, 87 percent experienced hypotension and 73 percent developed vascular leak syndrome or edema after high-dose treatment. Activation of CD25 on endothelial cells appears to contribute to these effects. Vascular leak syndrome induces hypovolemia that cascades into multi-organ dysfunction affecting the heart, kidneys, gastrointestinal tract and brain, producing reduced perfusion, ischemia, oliguria and neurological symptoms including confusion. In the pulmonary vasculature it can cause congestion and dyspnea, with severity increasing under continuous interleukin-2 therapy. The black box warning for high-dose interleukin-2 mandates intensive care monitoring due to the risk of acute kidney injury from renal hypoperfusion and cytokine-mediated neurotoxicity. Lifileucel, despite its efficacy, carries a treatment-related mortality rate of 7.5 percent, and its label restricts administration to trained clinicians at specialized facilities. The trial record is sobering: Creelan and colleagues reported two early deaths in patients with anti-PD-1-resistant metastatic lung cancer linked to rapidly progressive disease combined with the physiological stress of lymphodepletion and interleukin-2, and the IOV-LUN-202 trial in advanced non-small cell lung cancer was halted by the FDA in December 2023 following a treatment-related death.

Perhaps the most provocative findings concern interleukin-2’s subtler immunological sabotage. Work by Liu and colleagues and recent single-cell analyses show that interleukin-2 drives exhaustion of tumor-reactive CD8-positive T cells. In the absence of CD3/CD28 co-stimulation, interleukin-2 alone does not impair cytokine production but increases expression of the inhibitory receptors PD-1, LAG-3 and TIM-3. Prolonged high-dose exposure causes sustained STAT5 activation, which increases tryptophan hydroxylase-1 activity and enhances inhibitory receptor signaling, progressively degrading cytokine and effector molecule production. Single-cell transcriptomic and epigenetic studies trace the full exhaustion trajectory: prolonged interleukin-2 signaling pushes T cells from progenitor-like to terminally exhausted subsets, sequentially activating exhaustion-associated transcriptional programs such as TOX and NR4A while eroding memory-related gene expression. The tumor microenvironment compounds the problem, as hypoxia, nutrient competition, and immunosuppressive factors like TGF-beta and prostaglandin E2 interact with interleukin-2 to reinforce exhaustion-associated epigenetic programs. High CD25 expression also gives regulatory T cells a competitive advantage in capturing interleukin-2, allowing them to outcompete cytotoxic T cells and suppress the anti-tumor response, which in turn necessitates intensive lymphodepletion with its own toxicities.

The search for interleukin-2-independent TIL therapy has therefore become the field’s most active frontier. The clearest clinical success to date is GC101, developed by Shanghai Juncell Therapeutics as the first reported TIL therapy that eliminates post-infusion interleukin-2 altogether. GC101 is cultured in a pre-rapid expansion protocol medium containing interleukin-2, interleukin-7 and interleukin-15, followed by expansion with low-dose interleukin-2 at 300 international units per milliliter, allowing the final product to function without exogenous interleukin-2 after transfer. It uses a three-day lymphodepletion regimen of cyclophosphamide with hydroxychloroquine, and in phase 1 trials involving 14 evaluable patients with advanced gynecologic cancers achieved a 36 percent objective response rate, including complete remission in a patient with recurrent cervical cancer. Critically, GC101 can be administered entirely in general wards rather than intensive care units, improving safety and reducing costs. It has now advanced to a pivotal phase 2 randomized controlled trial in melanoma. Genetic engineering offers a parallel route: GC203, engineered to constitutively express membrane-bound interleukin-7, showed no grade 3 or higher non-hematologic adverse events in the first 18 ovarian cancer patients treated, while OBX-115 from Obsidian Therapeutics, expressing regulated membrane-bound interleukin-15, demonstrated in preliminary data from 10 patients an improved safety profile with no grade 4 or higher and only limited grade 3 non-hematologic events. Earlier attempts carry cautionary lessons, however: TILs engineered to secrete interleukin-12 achieved a 63 percent response rate in a 2015 first-in-human study but the trial was terminated early after life-threatening hemodynamic instability from interleukin-12 toxicity. CRISPR-based approaches are also emerging, with KSQ Therapeutics knocking out SOCS1 and Regnase-1 to enhance T cell activation, now in ongoing clinical trials.

The review’s authors are careful to separate proof-of-concept promise from validated clinical reality. Current engineered TIL evidence rests on small patient numbers, lacks randomized controlled trials, and involves short follow-up, while personalized genetic modification imposes stringent quality control demands and high production costs that limit scalability. Potential risks of sustained cytokine expression and CRISPR off-target effects demand careful trial design and long-term monitoring. Still, the strategic direction is clear. In the near term, the authors argue for predictive biomarkers incorporating T-cell exhaustion signatures, stem-like progenitor frequency and tumor microenvironment features to guide patient selection, alongside comparative studies to refine lymphodepletion regimens. Looking ahead, they call for next-generation interleukin-2 analogs that selectively stimulate effector T cells while sparing regulatory T cells, electroporation-based engineering as a safer alternative to viral vectors, and combination strategies with checkpoint inhibitors, targeted therapies or oncolytic viruses. If the field succeeds on both fronts, biologically precise interleukin-2-independent engineering and scalable, cost-effective manufacturing, TIL therapy could establish itself as a cornerstone of precision oncology for treatment-refractory solid tumors, finally resolving the paradox that interleukin-2 has imposed on cancer immunotherapy for nearly forty years.

Subject of Research: The dual role of interleukin-2 in tumor-infiltrating lymphocyte therapy for solid tumors

Article Title: Dual role of interleukin-2 in tumor-infiltrating lymphocyte therapy

Article References: Dual role of interleukin-2 in tumor-infiltrating lymphocyte therapy. (n.d.). https://doi.org/10.1007/s44272-026-00059-1

Image Credits: AI Generated

DOI: 10.1007/s44272-026-00059-1

Keywords: tumor-infiltrating lymphocytes, interleukin-2, TIL therapy, immunotherapy, solid tumors, lifileucel, melanoma, vascular leak syndrome, T cell exhaustion, engineered TIL, GC101, cancer immunotherapy