Tumour-infiltrating lymphocyte therapy landscape: prospects and challenges

Abstract

Tumour-infiltrating lymphocyte (TIL) therapy has emerged as a promising adoptive cell transfer strategy for solid tumours. The recent accelerated approval of lifileucel by the Food and Drug Administration marks a significant milestone in the clinical application of TIL therapy. This review comprehensively examines the historical development, biology, clinical efficacy, safety and limitations of TIL therapy. We explore advancements in TIL manufacturing, including novel culture techniques, genetic modifications and automation, to enhance scalability and effectiveness. Despite promising results, TIL therapy faces challenges such as high-dose interleukin-2 toxicity, complex manufacturing processes and immune evasion mechanisms. Emerging strategies, including checkpoint inhibitor combinations, engineered TIL constructs and metabolic reprogramming, aim to improve TIL therapeutic efficacy. This review provides insights into the evolving landscape of TIL therapy and its potential to enhance current cancer immunotherapy.

History of development and biology of tumour-infiltrating lymphocytes

On 16 February 2024, the Food and Drug Administration granted accelerated approval to lifileucel, a tumour-derived autologous T cell immunotherapy, for adult patients with unresectable or metastatic melanoma previously treated with immunotherapy or targeted therapies. This adoptive T cell transfer (ACT) immunotherapy is the first cellular therapy approved for a solid tumour. It involves isolation of tumour-infiltrating lymphocytes (TILs) from a patient’s resected tumours, which are then expanded to large numbers ex vivo using interleukin-2 (IL-2). Patients then receive lymphodepleting (LD) chemotherapy, followed by the infusion of TILs and then high-dose IL-2 to enhance the antitumour response. Approval of lifileucel was granted based on results of a multicentre, single-arm study in 153 patients that showed a durable response rate of 31.4%.¹

The discovery of the connection between lymphocytes and cancer dates back to 1863, when Rudolf Virchow observed lymphoid cells (leucocytes associated with inflammation) within cancerous tissues and hypothesised an association between chronic inflammation and the development of cancer.^{2 3} Cancer cells can invade across natural tissue barriers and disrupt normal tissue architecture, thus leading to initiation of inflammatory response. In this regard, the response of the body to a cancer may resemble body’s response to inflammation and wound healing.⁴

The idea of using the immune system to fight cancer originated in the 19th century. Wilhelm Busch and Friedrich Fehleisen first reported spontaneous regression of cancer following the development of erysipelas caused by Streptococcus pyogenes.⁵ Subsequently, William B. Coley treated patients who had sarcoma with heat-inactivated S. pyogenes and Serratia marcescens and observed the shrinkage of malignant tumours. This suggested that the immune system stimulated by infection can be directed to combat the cancer.⁶ In the 1950s, Nicholas Mitchison discovered that transferring lymphocytes led to rejection of tumour transplants in an animal model and subsequently provided the first report of ACT.7,9 However, in ACT, a large number of lymphocytes are required to suppress the tumour growth. The discovery of IL-2, known as the ‘T-cell growth factor,’ in 1976 revolutionised immunology research, making effective ACT possible. ^{10 11} Eberlein et al found that T cells expanded 8-fold to 10-fold over 7 days in IL-2 containing media and cured 93% of mice with syngeneic lymphoma.¹² Subsequent in vivo studies showed that concurrent administration of IL-2 with lymphocyte injection enhanced the effectiveness of these cells.¹³

TILs originate from haematopoietic stem cells in the bone marrow, which give rise to common lymphoid progenitors that migrate to the thymus for T cell development. In the thymus, T cells undergo T cell receptor (TCR) rearrangement, positive and negative selection, and lineage commitment into either CD4⁺ or CD8⁺ T cells.^{14 15} These mature naïve T cells enter peripheral circulation and become activated on encountering tumour antigens presented by dendritic cells in tumour-draining lymph nodes (TDLNs). Antigen recognition drives their clonal expansion and differentiation into effector T cells, some of which traffic to tumour sites, where they are termed TILs.16,18 The predominant component of TILs is CD8⁺ cytotoxic T lymphocytes, which recognise tumour antigens presented on major histocompatibility complex (MHC) class I molecules and mediate tumour cell killing through perforin and granzyme release, as well as the secretion of proinflammatory cytokines such as interferon-γ (IFN-γ) and tumour necrosis factor-α (TNF-α).¹⁹ CD4⁺ helper T cells are also commonly found in TIL populations and support antitumour responses by enhancing CD8⁺ T cell function and promoting immune activation.²⁰ Conversely, regulatory T cells (Tregs; CD4⁺CD25⁺FOXP3⁺) suppress effector T cell activity and are often enriched in tumours, correlating with poor clinical outcomes.²¹ Other functionally relevant subsets include natural killer (NK) cells, B cells and γδ T cells, which may contribute to cytotoxicity, antigen presentation or immune regulation.^{18 22} Recent studies have highlighted the role of stem-like CD8⁺ T cells, characterised by the absence of exhaustion markers such as CD39 and CD69, in supporting durable responses to adoptive TIL therapy.²³ TILs are recruited to tumours through chemokine-mediated trafficking, guided by gradients of tumour-secreted chemokines. A subset of six chemokines—CCL2, CCL3, CCL4, CCL5, CXCL9 and CXCL10—has been shown to be preferentially expressed in T cell–infiltrated melanoma tumours, acting through receptors such as CCR5 and CXCR3 on activated T cells.²⁴ TILs adhere to tumour endothelium via integrins (eg, LFA-1, VLA-4) interacting with adhesion molecules (eg, intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1)) on vascular endothelial cells, facilitating transendothelial migration.^{25 26} Once inside the tumour, TILs must infiltrate the stroma and interact with cancer cells, executing cytotoxic functions if not functionally suppressed. However, checkpoint molecule expression, immunosuppressive cytokines (eg, transforming growth factor-beta (TGF-β), IL-10) and metabolic constraints in the tumour microenvironment (TME) frequently impair TIL function and persistence.^{27 28} Figure 1 illustrates the recruitment process of lymphocytes to tumour.

Figure 1. Lymphocyte recruitment to tumour: Sequential pathway of T cell trafficking, localisation and infiltration into the tumour microenvironment. The diagram illustrates the sequential steps involved in lymphocyte trafficking and recruitment into the tumour microenvironment, beginning from their origin in the bone marrow to final localisation and infiltration into tumour tissues. T cell priming occurs in tumour-draining lymph nodes (TDLNs) after antigen presentation, followed by recruitment into the circulation and eventual migration across vascular endothelium into the tumour stroma. Chemokines and adhesion molecules mediate this process, facilitating T cell extravasation: adhesion, diapedesis and infiltration into tumour. CXCR, CXC motif chemokine receptor; ICAM-1, intercellular adhesion molecule-1; LFA-1, lymphocyte function-associated antigen-1; TIL, tumour-infiltrating lymphocyte; VCAM-1, vascular cell adhesion molecule-1; VLA-4, very late antigen-4.

Tumour mutational burden (TMB), the total number of somatic mutations in a tumour genome, correlates with the generation of neoantigens: non-self-peptides presented on MHC molecules and recognised by T cells. High TMB has been associated with improved responses to immunotherapies, including immune checkpoint blockade and adoptive T cell therapy, because it increases the probability of producing immunogenic neoantigens that can be targeted by TILs.^{29 30} Within the TIL population, neoantigen-reactive T cells represent the most potent effectors of tumour clearance, as they can specifically recognise and eliminate cancer cells while sparing normal tissues. Advances in single-cell RNA sequencing, TCR clonotyping and functional assays have enabled the identification and enrichment of neoantigen-reactive TILs for personalised TIL therapy.^{31 32} However, TILs are often a heterogeneous mixture of T cells, and many infiltrating clones may be bystander T cells—T cells present in the TME that are not specific for tumour antigens. These bystander TILs may be activated by viral or self-antigens and contribute little to antitumour activity.³³

Clinical activity of TILs

In 1980, Steven Rosenberg and colleagues described two different techniques for generating lymphocyte subpopulations: lymphokine-activated killer cells (LAK) and TILs. LAK cells, derived from peripheral blood lymphocytes cultured with IL-2, can lyse fresh tumour cells but not normal cells. TILs can be grown from tumours by culturing tumour cell suspensions with high-dose IL-2. In experimental tumour models, TILs proved to be 50–100 times more effective at killing tumour cells than LAK.^{34 35} In 1988, Rosenberg et al reported the first human trial demonstrating ACT efficacy in 20 patients with metastatic melanoma, who, after a single dose of cyclophosphamide, received TILs and IL-2. The overall response rate (ORR) was 60% in patients who were IL-2 treatment naïve and 40% in patients who received prior IL-2 treatment.³⁶ Use of cyclophosphamide to eliminate suppressor cells prior to TIL infusion³⁷ resulted in 35% ORR vs 31% ORR in those who were not treated with LD therapy in a study of 86 patients.³⁸ Tracked with radioactive indium-111, TILs localised to sites of metastatic deposits within 24 hours.³⁹ Addition of fludarabine to cyclophosphamide prior to TIL infusion led to improvement of TIL activity and resulted in 56% ORR, and 22% (from total of 93 patients), achieving a complete response (CR), and 19 maintaining a response beyond 3 years.40,42 Adoptive transfer of TILs recognising tumour antigen achieved an ORR of 46%.⁴⁰ The addition of 12 Gy total body irradiation (TBI) to LD chemotherapy achieved a 72% ORR.⁴¹

The need to shorten the time of TIL culture, usually lasting 6–8 weeks, led to the investigation of potency of unscreened for tumour reactivity, CD8+ enriched ‘young’ TILs. These TILs exhibit higher antigen reactivity, longer telomeres and increased expression of CD27 and CD28 receptors.43,45 Clinical activity of ‘young’ TILs was tested following non-myeloablative lymphodepletion (NMA) or combination of NMA and 6 Gy of TBI. In patients with treated only with NMA chemotherapy, 58% had ORR. Addition of TBI did not improve activity of ‘young’ TILs (48% ORR).⁴⁶ Besser et al reported that 72 out of 80 accrued patients had TIL cultures growing; however, 23 patients were withdrawn from the study because of clinical deterioration during TIL growth. This report was the first to report an intent-to-treat analysis (ORR of 29%). Shorter number of culture days (13.5 vs 17.5), higher number of infused cells (60×10⁹ vs 46×10⁹), higher number of CD8+ cells and faster recovery of absolute lymphocyte count 14 days after TIL infusion were associated with better disease control. The 3-year survival for all responders was 78%.⁴⁷ Treatment of 20 patients who received short-term cultured TILs resulted in 10% CR and 40% partial responses (PR), confirming that short culture TILs can generate active therapy, shorten wait for cell therapy product and thus decrease progression rate before TIL infusion.⁴⁸ In the pooled analysis of the Phase 2, C-144-01 study, lifileucel TIL therapy demonstrated an ORR of 31.4% in patients with advanced melanoma that progressed after treatment with immune checkpoint inhibitors (ICIs) and, if applicable, BRAF/MEK-targeted therapies. The application of TIL therapy has been tested in other solid tumour malignancies, such as renal cell carcinoma (29% ORR in 7 patients),⁴⁹ non-small cell lung cancer (23% ORR in 13 patients),⁵⁰ cervical cancer (44% ORR in 27 patients),⁵¹ non-cervical cancers associated with human papillomavirus (18% ORR in 11 patients),⁵² uveal melanoma (29% PR in 20 patients)⁵³ and osteosarcoma (36.7% ORR in 60 patients)⁵⁴ demonstrating activity of TIL treatment across different tumour histologies. A summary of contributions of selected clinical trials to the development of TIL is presented in table 1.

Table 1. Selected list of clinical trials exploring TIL therapy.

Author	Contribution to TIL therapy development	Conclusion
Rosenberg et al 199438	Demonstrated clinical efficacy of TIL therapy with high-dose IL-2 in 86 patients with metastatic melanoma. Better responses were observed in patients treated with TILs from younger culture, having shorter doubling time and having higher lysis of autologous tumour cells. Patients with TILs harvested from subcutaneous tumours rather than lymph nodes had better outcomes.	TIL therapy with high-dose IL-2 generated 34% ORR. There were no differences in response whether the patient received prior high-dose IL-2 therapy.
Schwartzentruber et al 1994170	Evaluated predictive markers of clinical response in 41 patients with melanoma.	Extranodal source of TILs, faster growing TILs and increased granulocyte macrophage colony stimulating factor secretions by TILs exposed to autologous tumour predicted better responses.
Dudley et al 200372	Method for generating melanoma-specific TIL cultures using high levels of IL-2 and rapid expansion protocols from 90 patients with melanoma.	Demonstrated efficient melanoma-specific TIL generation through optimised culture protocols.
Prieto et al 201045	Retrospective analysis of activity of CD8+ T cell enrichment method from young TIL cultures obtained from 92 patients with metastatic melanoma.	Simplified the methods of TIL generation with unselected, minimally cultured (‘young’) TILs.
Besser et al 2010 48	Clinical activity of minimally cultured, non-selected ‘young’ TILs combined with lymphodepletion and high-dose IL-2 in 20 patients with melanoma.	TILs derived from short culture time achieved 50% ORR.
Pilon-Thomas et al 2012171	Showed feasibility of TILs and high-dose IL-2 after lymphodepleting chemotherapy in 19 patients with metastatic melanoma; however, only 68% completed the therapy and were evaluable for responses.	26% ORR of metastatic melanoma in intention-to-treat patient population.
Radvanyi et al 2012117	31 patients with metastatic melanoma were treated with TIL therapy.	Tumour response was associated with higher number of TILs, higher proportion of CD8+ cells and higher frequency of CD8+ T cells expressing the B- and T-lymphocyte attenuator (CD272) protein. No significant difference in the telomere lengths in TILs between responders and non-responders was seen.
Besser et al 2013 47	Study analysed parameters predictive to response as well as the impact of prior anti-CTLA4 antibody or prior high-dose IL-2 on responses to TIL therapy.	Elevated above normal levels of lactate dehydrogenase, female gender, more days of TILs in culture and the lower total number of infused CD8+ cells were identified as independent negative predictive markers for clinical outcome. Six out of 11 patients with involvement of the CNS had intracranial responses.
Khammari et al 2014172	17-year median follow-up of patients participating in randomised trial to TIL/IL-2 therapy (44 patients) or IL-2 alone (44 patients) in adjuvant setting after resection of nodal recurrence of T1-4 melanoma.	There was no significant difference in OS between treatment groups, but in patients with lower tumour burden (one versus >more than one lymph node involved), TIL/IL-2 therapy was more effective than IL-2 alone. Better OS was seen with tumour-specific TILs and TILs with decreased Foxp3 expression.
Goff et al 201698	Randomised trial of NMA followed by TIL infusion or NMA and TBI followed by TILs in 93 patients.	Adding TBI to lymphodepleting chemotherapy did not improve CR or OS but introduced additional toxicities (such as thrombotic microangiopathy).
Chandran et al 201753	Phase 2 clinical trial of TIL in 21 patients with metastatic uveal melanoma. In 43% of patients the source of TILs was from liver, and in 57% was from extrahepatic metastases.	TIL therapy generated 29% PR in intention-to-treat population and CR in one patient with multiple liver metastases.
Mullinax et al 2018132	13 patients were evaluated to see if CTLA-4 blockade for four doses, starting 2 weeks before tumour harvest for TILs reduces attrition prior to TIL infusion.	Adding ipilimumab to TIL therapy was feasible and reduced rates of disease progression during waiting time for TIL expansion.
Nguyen et al 2019100	The study evaluated toxicity and activity of low-dose subcutaneous IL-2 (125 000 IU/kg/day, maximum 10 doses over 2 weeks after TIL infusion in 12 patients with metastatic melanoma.	TIL infusion followed by low-dose IL-2 had low toxicity rate with median number of IL-2 doses of 6.8, but induced only 16% PR.
Zhou et al 202054	Adoptive TIL therapy and anti-PD1 (nivolumab) in 60 patients with chemotherapy-resistant metastatic osteosarcoma.	Combined TIL and anti-PD1 therapy induced 36.7% ORR in metastatic osteosarcoma. Patients who received a higher number of TILs, CD8+TILs, a lesser number of CD8+PD1+TILs or CD4+FoxP3+TILs had improved PFS and OS.
Sarnaik et al 202156	Activity of lifileucel, an autologous, unmodified TIL therapy, in 66 patients with metastatic melanoma after failure of immune checkpoint and targeted therapy.	Lifileucel-induced durable responses in pretreated melanoma patients with ORR of 36%. Toxicity was related to NMA and high-dose IL-2.
Creelan et al 202150	Lung cancer tumours were harvested and subsequently expanded and tested for autologous reactivity. TILs were infused to 20 patients who experienced disease progression on nivolumab monotherapy.	35% of patients had tumour reduction in metastatic lung cancer refractory to initial nivolumab monotherapy. Neoantigen-reactive T cell clones persisted in peripheral blood after treatment.
Chesney et al 20221	Activity and safety of lifileucel in different cohorts, including cryopreserved TILs in 153 advanced melanoma patients with tumour progressing after failure of immune checkpoint inhibitor and BRAF/MEK inhibitors. Lifileucel was produced with a short 22-day manufacturing process.	Lifileucel-induced 31.4% durable ORR. The toxicity was related to NMA and high dose IL-2. It was resolved by day 14 after TIL infusion.
Rohaan et al 202255	Randomised clinical trial of TIL therapy (84 patients) vs ipilimumab (84 patients) for treatment of PD-1 refractory melanoma.	TIL therapy induced significantly longer PFS than anti-CTLA4 antibody in patients with melanoma progressing following PD-1 therapy.

CNS, central nervous system; CR, complete response; IL-2, interleukin-2; NMA, non-myeloablative lymphodepletion; ORR, overall response rate; OS, overall survival; PD-1, programmed cell death protein; PFS, progression-free survival; PR, partial responses; TBI, total body irradiation; TIL, tumour-infiltrating lymphocyte.

Long-term follow-up and durable responses

Rosenberg et al reported long-lasting clinical benefits, with patients obtaining CR surviving beyond 2 years.³⁸ Dudley et al documented 2-year survival rates of 30% in the trials with NMA not involving TBI.⁴¹ In the trial reported by Chesney et al, 41.7% of the responses lasted more than 18 months.¹ Furthermore, Rohan et al reported that in 75% of responders, the duration of response was longer than 66 months.⁵⁵

Patient-reported outcomes and quality of life

Patient-reported outcomes and quality of life (QoL) assessments were incorporated into a phase 3, multicentre, open-label trial, which randomised patients with stage IIIC or IV melanoma to TIL therapy or ipilimumab. QoL was evaluated at different timepoints using the European Organisation for Research and Treatment of Cancer Quality-of-Life Questionnaire Core 15 palliative care (EORTC QLQ-C15 PAL) scores. Patients who received TIL therapy had improved global QoL and physical and emotional functioning. In addition, they reported a lower symptom burden of fatigue, insomnia and pain than those in the ipilimumab group. However, a lower burden of nausea and vomiting was reported in patients treated with ipilimumab.⁵⁵

From patient selection to TIL expansion: a comprehensive overview of TIL therapy

Even though TIL therapy is effective against melanoma and other solid tumours, its application has limitations. Not all the candidates for TIL treatment would be able to receive TIL therapy. Multiple factors limit its broad implementation, including individual tumour heterogeneity, the baseline quality of TILs, the patient’s overall condition and immune status, the complexity of the manufacturing process, potential severe side effects, and issues of cost and accessibility. In the USA, patients who may be eligible for TIL therapy are usually referred to an authorised treatment centre. The complex process of TIL therapy is depicted in figure 2.

Figure 2. The process of tumour-infiltrating lymphocyte (TIL) therapy. The multidisciplinary team identifies optimal candidates for TIL, then surgical excision is planned synchronised with acceptance date of receiving manufacturing facility. During TIL production, patients are assessed for postsurgical recovery and for need of bridging therapy. Once the TIL products are available, patients are again evaluated for feasibility to undergo lymphodepleting therapy and high-dose interleukin-2 (IL-2). When TIL products arrive at hospital, patients are admitted, where they receive lymphodepleting chemotherapy. TIL products are thawed at the bedside and then given intravenously on day 0. A few hours later, high-dose IL-2 injections start and are given every 8–12 hours for up to six doses. Patients are then monitored for recovery and discharged to be followed in an outpatient setting.

Patient selection, tumour resection and tissue collection

Selecting appropriate patients for TIL therapy is a multifactorial decision, influenced by tumour type, disease stage, prior treatments, overall health status and the ability to obtain TILs from resected tumour tissue. Medical selection of patients considers their ability to safely undergo surgical tumour resection, LD chemotherapy, TIL infusion and high-dose IL-2. Hence, patients typically must have an Eastern Cooperative Oncology Group (ECOG) performance status of ≤1; age less than 75; absence of severe comorbidities such as irreversible cardiac dysfunction, impaired pulmonary function or autoimmune disorders requiring immunosuppression. Patients with brain metastasis are often excluded.55,57

Surgical oncology evaluation is an early step in the selection and ensures suitability of the patient with respect to perioperative risks of surgery as well as identification of sufficient tumour tissue for manufacture of a TIL product.

The surgical approach should be chosen to minimise healing time to ensure patients are ready for LD chemotherapy as soon as a TIL product is received, as well as for possible bridging therapy in the interval after surgery. Prospective clinical trials have included patients with good performance status, which may be a challenge in real-world settings.⁵⁶ The pace of progression of disease must be considered and surgical evaluation may help to identify impending issues that may require intervention prior to proceeding with LD, such as biliary obstruction or ulceration of soft tissue tumours. Perioperative medical therapies must be considered both for the impact on surgical morbidity and the impact on TIL abundance and function.

Patients may have ongoing steroid treatment related to toxicity from prior use of ICI therapies. Steroids will impair abundance and effector function of TILs.⁵⁸ While there is limited guidance on optimal timing of tissue procurement relative to steroid dosing, our practice has been to avoid tissue procurement within 2 weeks of supraphysiologic doses of steroids.

The effect of antineoplastic systemic therapies for cancer on TIL manufacturing is unclear. Neoadjuvant immunotherapy has not been shown to have a detrimental effect on surgical risk.⁵⁹ The effects of targeted B-Rapidly Accelerating Fibrosarcoma (BRAF) and Mitogen-activated Protein Kinase Kinase (MEK) therapy on immune function are not well defined, with mixed reports on both modest immunosuppressive and immunostimulatory effects.^{60 61} BRAF inhibitors are generally held for 1–3 days prior to surgery, and MEK inhibitors 5–7 days prior to surgery, with resumption 5–7 days after surgery due to negative effects on wound healing and angiogenesis.^{62 63}

Preoperative cytotoxic chemotherapy is likely associated with a negative impact on TIL generation^{64 65} and is associated with surgical complications.⁶⁶ Therefore, exclusion of chemotherapy within at least 3 weeks before and after surgery is generally practised.

Tumour selection has a significant impact on TIL manufacture. Larger tumour volume is associated with more active TILs, possibly due to larger and more diverse initial population of TILs.⁶⁴ Different cut-off values for tumour size have been described with current commercial minimum tumour volume to be 1.5×1.5×1.5 cm, which may be achieved with resection of multiple tumours as needed.⁵⁶ At present, this volume of tumour is only achieved with surgical resection. TIL isolation and expansion have been described with percutaneous biopsy, although with lower rates of successful in vitro expansion and total viable cell count.⁶⁷ Patients with tumours that fail to manufacture viable and functional TIL can undergo another resection and TIL manufacture with similar rates of success to initial surgery.⁶⁴

Different organ sites may be associated with differential TIL yield and activity. Data do not suggest a difference in T-cell receptor (TCR) repertoire based on site of tissue procurement.⁶⁸ Lymph node, subcutaneous and soft tissue, visceral and pulmonary sites have all yielded suitable rates of successful culture and activity.5764 65 69,71 Sites that may have significant bystander lymphocyte populations such as spleen may be less favourable, and limited data suggest visceral and particularly hepatic sites may also be less favourable though not prohibitive.⁶⁴ Care should be taken to avoid tumours where necrotic tissue comprises a significant portion of tumour volume. Grossly contaminated sites such as ulcerated soft tissue or the mucosal surface of the aerodigestive tract should also be avoided due to risk of contamination during the manufacturing process. Tumours previously treated with locoregional therapies including radiation and intralesional therapies should be avoided.

Preoperative imaging is essential to identify suitable tumours. Care should be taken to assess necrosis, particularly in rapidly growing masses or patients receiving ongoing therapy. Biopsy should be performed preoperatively in instances where confirmation of tumour is required. Intraoperative pathology may also be performed as necessary to confirm viable tumour.

Lymph node and superficial sites should generally be favoured over visceral locations. Minimally invasive surgery should be pursued if available to help minimise recovery time. Complex surgeries with elevated perioperative risk or requiring significant reconstruction or healing time should be avoided.⁶⁸

Sterile prosection is used to isolate viable tumour tissue from any excess non-tumour tissue or areas of gross necrosis.⁷² In general, portions from the periphery of the mass or adjacent to blood vessels or lymphatics are preferred to ensure viable tissue and infiltration of TILs and antigen presenting cells.⁶⁸ Mincing the tumour at the time of prosection is not required and may result in further fragmentation or dissociation of tissue preventing subsequent evaluation and plating for culture. After prosection, the aggregate of viable tumour tissue should measure up to 4×4×4 cm, but not less than 1.5×1.5×1.5 cm.⁷² The tumour tissues are placed in a sterile media and maintained at 2–8°C in a refrigerator until pick up by prearranged courier for shipment to the manufacturing facility.⁷³

When a portion of tumour is required to be submitted for pathology or research, sterile prosection should occur in the operating room. Care should be taken to minimise excision of tumour specimen for non-manufacturing purposes to maximise the amount of tumour submitted for manufacture. Surgeons should document tumour location/tissue type, size, qualitative description and weight.⁷⁴ Postoperative follow-up should be performed within 2 weeks of surgery to ensure adequate recovery and to evaluate for other emerging issues that may impact timeline for LD.

Successful tissue procurement requires adherence to established good manufacturing practice, chain of custody and of identity protocols. In the USA, tumour tissue procurement protocols and oversight will likely be established in the forthcoming eighth edition standards of the Foundation for the Accreditation of Cellular Therapy.

Isolation and expansion of TILs

Initial stages of T cell outgrowth from tumour tissue differ depending on if processing is done in the same institution or must be shipped outside. When TIL isolation is done near the site of tumour tissue collection, tumour tissue is transferred fresh to the local laboratory. Tumour tissue processing occurs promptly, usually within 1–3 hours, to minimise cellular degradation.⁷⁵ Tumour fragments are cut into small pieces (1–3 mm³) to increase surface area and enhance lymphocyte extraction.⁷⁶ When tumour tissue must be shipped, it is preserved in specialised media, such as HypoThermosol, at controlled temperatures (2–8°C) before processing to maintain viability.⁷⁷ Usually, antibiotics are added to media to prevent bacterial or fungal growth.

Establishing single-cell suspensions from the tumour tissue is the next step. This step is critical for liberating TILs from the TME. In many protocols, tumour tissues are enzymatically digested into single-cell suspensions using enzymes such as collagenase, hyaluronidase and DNase.^{72 78} Alternatively, a mechanical method using a cell homogeniser may be used to achieve single-cell suspensions.⁷⁹

Single-cell suspensions are diluted and distributed into individual wells of 24-well culture plates containing IL-2-enriched media (3000–6000 IU/mL). Typically, 48 wells (two 24-well plates) are used for pre-rapid expansion phase (REP) culture.^{78 80} Wells are monitored every 3–4 days for feeding or splitting as needed. Over time, residual tumour cells die since the culture conditions favour lymphocyte growth. It takes 3–5 weeks for TILs to outgrow tumour cells and reach the desired 60×10⁶ viable TILs for the subsequent REP stage. At that point, TILs can be screened for tumour recognition by IFN-γ release assay. Reactive TILs are then pooled and expanded during the REP using IL-2 (3000 IU/mL), agonistic anti-CD3 (30 ng/mL) and irradiated allogeneic peripheral blood mononuclear cells (PBMC) feeder cells at a 1:200 TILs-to-PBMC ratio. On day 14, the cells are harvested, concentrated, washed and prepared for administration. T175 flasks and gas-permeable cell culture bags are used for REP. From tumour resection to the final TIL production, the process described above generally takes 5–7 weeks.

Advancements in expansion and manufacturing techniques

The increased expansion rates have significantly improved, and the culture period has decreased to 14 days with the addition of agonistic anti-4-1BB antibody to the pre-REP phase.⁸¹ Another development was the addition of monoclonal antibodies activating T cells, such as anti-CD3 and anti-CD28, to growth media in combination with feeder cells to stimulate TILs and enhance their proliferation in REP phase.^{82 83} Another innovation is the use of artificial antigen-presenting cells as an alternative to traditional allogeneic PBMC feeders. Use of the human erythroleukaemic K562 cell line modified to express costimulatory molecules such as CD86, 4-1BBL, CD64 and membrane-bound IL-15 as a feeder layer demonstrated comparable TIL expansion rates while providing a more scalable and consistent approach.⁸⁴

The manufacturing process for TIL therapy has also witnessed numerous innovations aimed at improving efficiency, scalability and compliance with Good Manufacturing Practice (GMP) standards. Replacement of culture flasks and bags with functionally closed systems such as G-Rex100M-CS culture flasks and a GatheRex Cell Harvest pump at the REP phase has not only increased the production of viable TILs but also reduced the number of vessels, media consumption, incubator space and labour compared with traditional culture methods. Other closed-system bioreactors, such as the ReadyToProcess WAVE,^{85 86} Xuri Cell Expansion System⁸⁷ bioreactors and Lovo processing system, also provide a controlled environment for large-scale TIL expansion while minimising contamination risk.⁸⁸ For example, the Wave bioreactor supports the continuous influx of fresh medium, creating an optimal environment for TIL growth.⁸⁹ These innovations not only ensured GMP compliance but also reduced variability in TIL culturing.⁹⁰ Optimised TIL culture expansion process has ensured consistent production of the optimal therapeutic dose of 60×10⁹ cells for all patients, compared with achieving this dose in only 40% of patients with traditional culture methods.⁸⁰

Quality and quantity of TILs: optimising the therapeutic product

Advanced flow cytometry techniques, combined with TCR sequencing, are now used to evaluate more precisely TIL quality. Functional assays, such as cytotoxicity tests, are used to measure the ability of TILs to kill tumour cells in vitro, ensuring that only the most potent lymphocytes are selected for infusion. The phenotype of TILs is also evaluated using flow cytometry, focusing on markers such as CD8+, CD4+ and PD-1. In terms of quantity, the goal is to achieve cell count, typically ranging between 1×10⁹ and 2×10¹¹ TILs. Throughout the process of TILs production, quality checks are performed. In the pre-REP phase, assessments for sterility, viability, total cell count and tumour-specific T cell activation are performed. At the time of TILs release after the REP phase, the product is tested for mycoplasma contamination, sterility, purity, identity, viability, cell count and tumour-specific T cell activation.⁹¹

Bridging therapy in TIL treatment: clinical considerations

Bridging therapy is employed to manage tumour progression in patients while TILs are being expanded. Given that TIL expansion can take several weeks,⁵⁶ during which time the tumour may continue to grow, bridging therapies are used to stabilise disease and create a favourable immune microenvironment. Common strategies include chemotherapy, radiation therapy and targeted therapy, each selected based on the patient’s tumour type, clinical condition and prior treatments.⁹² Following the failure of ICIs, tumour-reactive lymphocytes are still present within tumours, and chemotherapy may not only directly reduce tumour burden but can also prime the tumour immune environment.^{93 94} Additionally, radiation therapy may be employed to control local disease progression while awaiting TIL expansion.⁹³ Radiation therapy, when combined with ICIs, has been shown to synergistically boost immune infiltration into tumours.⁹⁰ The median time from tumour resection to TIL infusion typically ranges from 4 to 8 weeks; in one study where TILs were derived from non-small cell lung cancer, patients had to wait for TIL infusion median of 35.5 days after tumour resection.⁹⁵ In tumours that have actionable mutations, use of targeted therapy before TIL harvest and during TIL production was safe, induced responses, promoted activity of TILs against low frequency clonotypes and decreased attrition.⁹⁶

Lymphodepletion in TIL therapy

Lymphodepletion is a preparatory step in TIL therapy, aimed at creating an optimal immune environment for the infusion of TILs. This process reduces the patient’s regulatory T and other immunosuppressive cells, which can impede the engraftment and efficacy of the infused TILs. Additionally, reduction in number of immune cells before TIL infusion may decrease cytokine release syndrome when IL-2 is administered. Earlier initiation of lymphodepletion (eg, day −7 rather than day −5) led to improved outcomes.⁹⁷ The commonly used lymphodepletion regimens involve the administration of cyclophosphamide and fludarabine. Cyclophosphamide, administered at 60 mg/kg/day for 2 days, is typically followed by fludarabine at 25 mg/m²/day for 5 days. These agents effectively deplete endogenous lymphocytes, creating the necessary ‘space’ for TILs to expand in vivo after infusion. This combination has been critical in achieving durable clinical responses and enhancing TIL persistence in patients. The addition of total body irradiation (TBI) improved lymphodepletion outcomes in some studies, but it carried a higher risk of toxicity and is no longer used.⁹⁸

TIL infusion

The infusion of TILs marks the culmination of weeks of TIL expansion and conditioning. Once lymphodepletion is complete, TILs are administered intravenously.⁷² The TILs are typically resuspended in a saline solution with human albumin for infusion, typically 30–60 min per bag.⁹⁹ Continuous monitoring is essential to detect any immediate adverse effects, such as infusion reactions or cytokine-related toxicities. After a short observation period following TIL infusion, high-dose IL-2 administration starts to support the in vivo proliferation and survival of the infused TILs.⁵⁶

IL-2 in TIL therapy: role and toxicity management

High-dose IL-2 plays a fundamental role in TIL therapy by supporting the proliferation and survival of infused TILs. Historically, IL-2 was administered at 600 000 to 720 000 IU/kg every 8–12 hours from up to 6 to 14 doses, but was reduced to only six doses in recent protocols.^{46 56} IL-2 administration is associated with significant toxicities, including capillary leak syndrome, fever, hypotension and organ dysfunction, often requiring cardiac and intensive care unit monitoring during administration, limiting clinical use of TILs to patients without cardiovascular, pulmonary and renal comorbidities.^{55 64} Several strategies are employed to manage these toxicities, including making care decisions regarding that next dose of IL-2, monitoring of symptoms, as well as supportive care for complications such as febrile neutropenia and thrombocytopenia.⁷³

A decrescendo IL-2 dosing regimen has been explored to reduce the severity of side effects while maintaining TIL efficacy with preserved long-term complete responses.⁹¹ Low-dose subcutaneous IL-2 was also tested (125 000 IU/kg/day, maximum 9–10 doses over 2 weeks), but this resulted in only 16% PR.¹⁰⁰ Alternative strategies to IL-2 are emerging, such as modification of TILs to express inducible with acetazolamide membrane-bound IL-15.⁸⁹

Post TIL treatment monitoring

After completion of IL-2 therapy, the patient stays in the hospital until the toxicities from LD and IL-2 therapy improve. Granulocyte colony stimulating factor is used to shorten the time of severe neutropenia, and patients receive red blood cell and platelet transfusion, if needed for severe anaemia and thrombocytopenia. Patients should be close to the treatment facility for 14 days to monitor and treat delayed toxicities. The use of steroids is contraindicated. Four to 6 weeks after TIL infusion, restaging imaging studies are performed to assess the activity of the therapy. Starting with the initiation of lymphodepletion, prophylactic antibacterial, antifungal and antiviral therapy is given. While antibacterial and antifungal therapy stops with neutrophil recovery, prophylaxis against Pneumocystis jirovecii pneumonia continues until the CD4 lymphocyte count is >200/mm³. Antiviral prophylaxis and revaccination programmes follow established guidelines borrowed from experience of autologous haematologic stem cell transplantation.^{101 102}

Resistance to TIL therapy: mechanisms and contributing factors

Resistance to TIL therapy remains a significant hurdle, limiting its efficacy and broader application. Resistance to TIL therapy is multifactorial, including alterations within tumour cells, dysfunctions in T cells and immunosuppressive elements of the TME. Understanding these mechanisms is crucial for developing strategies to enhance the effectiveness of TIL therapy.

Tumour-intrinsic factors

Tumour-intrinsic resistance arises from alterations within cancer cells that enable them to evade immune detection and destruction. For instance, tumour cells can downregulate or mutate key components of antigen presentation, such as MHC class I molecules, thereby escaping recognition by T cells.103,106 Additionally, tumour evolution and clonal heterogeneity often give rise to cell variants that lack target antigens recognised by TILs.^{99 107} Alteration of oncogenic signalling pathways, such as MAPK, PI3K/PTEN, WNT/β-catenin and IFN-γ, can promote immune resistance.108,110 Activation of the WNT/β-catenin pathway has been shown to suppress chemokine expression necessary for T cell infiltration into tumours.¹¹¹ Defects in the STING (stimulator of interferon genes) pathway also result in immune evasion to TIL therapy.¹¹² Tumour cells may upregulate immune checkpoint ligands such as PD-L1, which bind to inhibitory receptors on T cells, leading to reduced T cell activation and function.¹¹³

T cell factors

One of the primary challenges in TIL therapy is T-cell exhaustion. Prolonged exposure to tumour antigens drives T-cell exhaustion, marked by sustained expression of inhibitory receptors such as PD-1, LAG-3 and TIM-3 on T cells, along with diminished effector functions.114,116 Within the TME, TILs frequently display this exhausted phenotype, evidenced by reduced production of key effector molecules like IFN-γ, TNF-α and IL-2.⁸⁴ Impaired mitochondrial dynamics and metabolic dysfunction further hinder TIL proliferation and cytokine secretion, contributing to T-cell exhaustion.^{82 83} Additionally, elevated CD36 expression on CD8⁺ T cells has been associated with T-cell exhaustion and a dysfunctional TME.⁸⁸

Variability in the proliferative capacity and phenotypic quality of TILs during ex vivo expansion may lead to suboptimal TIL expansion and limit their effectiveness on reinfusion.¹¹⁷ Restricted TCR clonality among TILs may also impair their ability to target heterogeneous tumour antigens.¹¹⁸

TME factors

TME is the complex ecosystem of cells, molecules and blood vessels that surround and support a tumour. Tumours actively recruit immunosuppressive cells such as Tregs, myeloid-derived suppressor cells and tumour-associated macrophages. These cells inhibit effector T cell function and release suppressive cytokines (eg, TGF-β, IL-10) and deplete nutrients (eg, arginine, tryptophan) essential for T-cell function.119,122

TGF-β polarises immune cells towards suppressive phenotypes and inhibits T cell activation.¹²³ TME is typically characterised by hypoxia, low pH and nutrient depletion due to aberrant vasculature and rapid tumour cell proliferation. These conditions impair TIL metabolism and function.^{82 124 125} Dense extracellular matrix components and abnormal tumour vasculature can physically impede the infiltration of TILs into tumour sites.¹²⁶

There are emerging studies which suggest that cytokines, such as IL-7¹²⁷ and IL-15,¹²⁸ may reverse TIL exhaustion, especially during the early stages of TIL culture. These cytokines help maintain TIL functionality and promote the development of memory T cells, which are crucial for long-term antitumour immunity.⁸⁸

Resistance to TIL therapy is a complex and multifaceted challenge involving tumour-intrinsic changes, T cell dysfunctions and TME influences. Addressing these barriers requires a comprehensive approach, including combination therapies targeting multiple resistance pathways, genetic modification of TILs to enhance their resilience and modulation of the TME to support effective immune responses.

Limitations and challenges

Despite the promising outcomes associated with TIL therapy, significant challenges remain. The toxicity associated with high-dose IL-2 and the labour-intensive process of TIL expansion pose substantial barriers to its widespread clinical adoption.

IL-15 has emerged as a promising alternative to high-dose IL-2 in enhancing TIL efficacy with reduced systemic toxicity. For instance, Shen et al introduced an engineered IL-15 receptor fused to anti-programmed cell death protein (PD-1) (αPD-1-IL-15-R), which selectively targets PD-1+CD8+ T cells within the TME. This approach minimises peripheral NK cell toxicity and enables targeted cis-delivery, resulting in significant tumour-specific CD8+T cell expansion and antitumour effects with minimal systemic side effects.¹²⁹ Xu et al further developed this concept using an engineered IL-15 mutein fused to anti-PD-1, which enhanced intratumoral T-cell function and overall antitumour immunity.¹³⁰

Combination therapies

The integration of TIL therapy with ICIs has shown remarkable potential in amplifying antitumour responses by addressing immune exhaustion and resistance mechanisms. For instance, TIL therapy with anti-PD1 antibody, pembrolizumab in immunotherapy-naive patients with advanced melanoma resulted in a 65.2% ORR and a 30.4% CR rate.¹³¹ When the anti-CTLA4 antibody, ipilimumab, was given before tumour harvest for TIL generation, and then an additional cycle of ipilimumab was given followed by lymphodepletion, TIL infusion and high-dose IL-2, a 38.4% ORR was observed, with 30% lasting beyond 1 year.¹³²

Modification of TILs

Modification of TILs offers a promising strategy to overcome challenges such as antigen specificity, immunosuppressive TME, TIL persistence within tumours and IL-2 toxicity. OBX-115, an IL-2-sparing TIL product engineered to express membrane-bound IL-15,¹³³ has demonstrated enhanced efficacy with 100% disease control rate (nine patients), and a favourable safety profile in patients with ICI-resistant metastatic melanoma.¹³⁴ Knock down using siRNA of transcriptional regulator ID3 (inhibitor DNA binding 3), which is expressed mostly in exhausted CD8 cells, resulted in significantly increased killing of lung adenocarcinoma cells.¹³⁵ Cytokine-inducible SH2-containing protein (CISH) in CD8+T cells silences TCR signalling and inhibits T cell expansion and activity. Deletion of CISH restored antitumoral activity of CD8+ T cells.¹³⁶ The strategy of knockout of CISH with CRISPR-Cas9 in TILs was tested in patients with metastatic colorectal cancer and resulted in 50% ORR (six patients).¹³⁷ Suppressor of cytokine signalling 1 (SOCS1) inhibits T cell signalling induced by cytokines, impedes increase in the number of central memory cells, and intermediate and effector exhausted T cell subsets. CRISPR-Cas9 engineered SOCS1 TILs (KSQ-001 manufactured product) had increased responses to cytokines and antitumour activity in mice.¹³⁸ Knocking out with CRISPR-Cas9 of PD-1 in TILs resulted in 87.53% reduction in cell surface PD-1 expression.¹³⁹ CRISPR PD-1 edited TIL therapy is being tested in melanoma (NCT06783270). Innovative platforms like CoStAR, a CD28/CD40 chimeric costimulatory receptor, provided synthetic costimulation that enhanced TIL functionality, persistence and antitumour activity. By delivering tumour-specific costimulatory signals only in the presence of TCR engagement, CoStAR augmented cytokine production and cytolytic activity of TILs while reducing dependency on exogenous IL-2.¹⁴⁰ Targeting metabolic checkpoints like sirtuin-2 enhances TIL functionality within the TME, producing more resilient T cells capable of withstanding metabolic pressures.¹⁴¹ Enhancing activity of TILs towards subdominant tumour antigens through stimulation of Toll-like receptor in TLR2-MyD88 TILs resulted in improved T-cell infiltration, reduced T-cell exhaustion, longer survival, cytokine production, expansion and cytotoxicity.^{142 143} Expression of the cytolytic granule-associated molecule natural killer cell granule protein-7 (NKG7) improved the antitumour activity of murine tumour antigen-specific CD8+ T cells.¹⁴⁴ Loss of NR4a1 and NR4a2 (nuclear receptor subfamily four group A) CD8+TILs increased TCF1+ stem-like precursors, promoted glycolysis and oxidative phosphorylation, and enhanced TIL persistence in the TME.¹⁴⁵ Stem-like CD8⁺ T cells, characterised by the absence of exhaustion markers such as CD39 and CD69, were associated with durable responses to adoptive TIL therapy.²³ The pharmacological inhibition or CRISPR knockout of PI3K and AKT resulted in an increase in the population of activated effector CD8+TILs, an increase in CD39- CD69- memory T cells, a TCF1 expression maintaining T cell stemness, increased IFN-γ and TNF-α production, and resulted in intensified lysis of patient-derived tumours.¹⁴⁶ Tetraspanin molecules like CD81 and CD82 act as co-stimulatory regulators, augment expression of activation markers (CD25, CD69) and support differentiation of T cells into long-lived central memory subsets (CCR7⁺ CD45RA⁻) through sustained IL-2 signalling and transcriptional changes.¹⁴⁷ TMEM123, glycosylated mucin-like protein, expression in CD8+TILs facilitated TIL activation, migration, diapedesis and clustering within tumours.¹⁴⁸ Reprogramming TILs with Sendai virus vectors into iPSC has paved the way for generating a renewable source of tumour-reactive T cells that retain pluripotency and hold potential for long-term use.¹¹⁸

Overcoming the TME challenges

Within the TME, tumour-infiltrating immune cells, such as cancer-associated fibroblasts,¹⁴⁹ tumour-associated macrophages,¹⁵⁰ myeloid-derived suppressor cells,¹⁵¹ plasmacytoid dendritic cells¹⁵² and tumour-infiltrating regulatory T-cells,¹⁵³ contribute to an immunosuppressive environment and represent a challenge to effective TIL therapy. Current efforts to address this challenge include blockade of CCL2/CCR2 or CSF-1/CSF-1 receptor pathways to decrease infiltration of tumour by tumour-associated macrophages¹⁵⁴ or influencing polarisation of macrophages to an inflammatory M1 phenotype.¹⁵⁵ Strategies to target immunosuppressive myeloid cells are also being developed.¹⁵⁶ TME contains immunosuppressive chemokines¹⁵⁷ and cytokines, such as TGF-β, IL-4, IL-5, IL-6 and IL-13.^{158 159} A TGFBR2 knockout in TILs rendered them resistant to immunosuppressive TGF-β signalling.¹⁶⁰ HLA-E+positive tumour cells suppress the cytotoxic function of TILs. This can be overcome by knocking out NKG2a. This effective strategy was reported when using Vδ2 cells (subtype of innate-like γδ T cells).¹⁶¹ Lastly, the spatial organisation of TILs within the TME, such as their clustering in brisk immune phenotypes, is associated with better immune responses and outcomes. Optimising TIL localisation and activation through cytokine production and co-stimulatory signalling can further mitigate the suppressive effects of the TME.¹⁶²

Identification of functional and tumour-reactive TILs

Presence of intracellular CD137 and production of TNF and IFN-γ can be quickly identified by flow cytometry TILs that are tumour reactive.¹⁶³ Chronic TCR stimulation during TIL expansion preserves the phenotypic and transcriptomic characteristics of tissue-resident T cells while reducing exhaustion markers, such as PD1 expression.¹³⁵ Use of AKT or PI3K inhibitors during REP has enhanced the generation of effector CD8+T cells with higher activation markers and memory characteristics.¹⁴⁶

Broadening applicability to other tumour histologies

Clinical trials, such as IOV-COM-202, have included cohorts with ovarian, renal and other solid tumours, demonstrating the feasibility of TIL therapy in these malignancies. Notably, other than in melanoma, significant activation of tumour-specific TILs was demonstrated in ovarian cancer and sarcoma.¹³¹ Comprehensive immune profiling has revealed distinct infiltration patterns across 33 cancer types. Tumours such as renal cell carcinoma and microsatellite instability-high cancers exhibit high CD8+T cell infiltration, making them suitable candidates for TIL therapy.¹⁶⁴ Moreover, pan-cancer analyses have identified shared tumour-reactive T cell states, such as terminally exhausted T-cell and follicular helper T cell/T helper 1 dual-functional T cell subsets, across diverse cancers, including liver and oesophageal carcinoma, reinforcing the potential of TIL therapy in non-melanoma settings.¹⁶⁵ In highly immunosuppressive tumours like pancreatic ductal adenocarcinoma, TIL populations with robust tumour reactivity have been successfully expanded, demonstrating feasibility even in challenging contexts.¹⁶⁶ In ovarian and cervical cancers, inhibiting pathways like AKT enhances TIL propagation and memory-like phenotypes.¹⁴⁶ Innovative technologies, such as the CoStAR platform, have broadened the scope of TIL therapy by improving reactivity in tumours expressing folate receptor alpha, such as lung, renal and ovarian cancers.¹⁴⁰

Precision-engineered TIL

The identification of HLA-C*08:02–restricted T-cell receptors targeting KRAS G12D neoepitopes has demonstrated the potential of TIL therapy in recognising driver mutations in cancers like pancreatic and colorectal cancer.¹⁶⁷ Another illustration of effective TIL therapy was described in a patient with hormone receptor negative breast cancer where reactive TILs were selected against mutated proteins SLC3A2, KIAA0368, CADPS2 and CTSB.¹⁶⁸ Similarly, using epitope prediction pipelines has allowed the generation of neoantigen-directed TIL products with superior tumour growth control compared with standard protocols, supporting the clinical feasibility of personalised approaches.¹⁶⁹

Conclusion

TIL therapy has been shown to be an effective strategy for the treatment of refractory melanoma and is being developed for other solid tumours. While TIL therapy holds a lot of promise for long-term control of metastatic solid tumours, further developmental and clinical work needs to be done to overcome shortcomings of TIL therapy.

Data availability statement

No data are available.