Discrepant CD3+ TILs in PD-L1–Negative NSCLC: Favorable Outcome in an Elderly Patient Treated With Nivolumab, Ipilimumab, and Chemotherapy

Introduction

The tumor microenvironment is now considered to play a crucial role in cancer pathogenesis and in determining the efficacy of anti-cancer therapies. Key components of this environment include immune cells, stromal cells, blood vessels, and the extracellular matrix [1]. Among these, tumor-infiltrating lymphocytes (TILs) have gained increasing attention in cancer immunology [2,3], as the therapeutic effects of ICIs are thought to be mediated by the reactivation of TILs through blockade of the PD-1, PD-L1, and/or CTLA-4 pathways.

Signal transduction through programmed cell death-1 (PD-1), programmed cell death-ligand 1 (PD-L1), and cytotoxic T lymphocyte antigen-4 (CTLA-4) attenuates T lymphocyte activity, which is essential for immune surveillance of cancer cells. Immune-checkpoint inhibitors (ICIs) targeting these molecules can restore T cell function.

Remarkable advances in ICIs have led to a paradigm shift in cancer treatment, offering the potential for durable responses – even in advanced disease – when used as part of multimodality approaches. Although PD-L1 expression is commonly employed as a biomarker to guide ICI therapy, recent clinical trials and case reports suggest that selected patients with PD-L1–negative non-small cell lung cancer (NSCLC) may still benefit from ICI-based regimens, particularly those incorporating anti-CTLA-4 antibodies and cytotoxic chemotherapy. For example, previous case reports have documented non-small cell lung cancer (NSCLC) patients who had favorable responses to chemoimmunotherapy despite negative PD-L1 tumor proportion scores, indicating that factors other than PD-L1 expression may influence treatment efficacy [4,5].

Here, we report a case of advanced squamous cell carcinoma of the lung with a PD-L1 tumor proportion score (TPS) below 1%. The patient showed a remarkable response to treatment with anti-PD-1 and anti-CTLA-4 monoclonal antibodies in combination with carboplatin and paclitaxel. Pathological examination prior to treatment revealed numerous CD3-positive lymphocytes infiltrating the tumor and its periphery. CD4- and CD8-positive TILs were present in approximately equal numbers. These findings raise the hypothesis that TILs exert meaningful antitumor activity independent of the PD-1/PD-L1 axis, even in tumors lacking PD-L1 expression.

Case Report

In August 20XX, an 80-year-old man presented with a cough and was referred to our hospital for detailed evaluation of abnormal chest X-ray findings (Figure 1A). Chest computed tomography (CT) revealed a 4.12×7.24×6.68 cm mass in the left lower lobe (Figure 2A), with mediastinal lymphadenopathy and bilateral pleural effusions. His smoking history was 1 pack per day for 50 years. Physical examination demonstrated decreased respiratory sounds in the left lower lung field due to the primary cancer and associated pleural effusion. His Eastern Cooperative Oncology Group performance status (ECOG-PS) was 0. Enhanced head magnetic resonance imaging (MRI) detected no brain metastases.

Fluorodeoxyglucose positron emission tomography (PET)/CT revealed the clinical stage was IVB (cT4N3M1c) due to metastases involving both contralateral and supraclavicular lymph nodes as well as multiple extra-thoracic lymph nodes (Figure 1B). Serum levels of various tumor markers were measured; cytokeratin 19 fragment (CYFRA) 52.4 ng/ml (reference range <3.5 ng/ml). Except for CYFRA, the patient’s laboratory findings were unremarkable. He underwent transbronchial lung biopsy, yielding a diagnosis of squamous cell carcinoma (Figures 3, 4A) without PD-L1 expression (TPS less than 1%) (Figure 4B). Driver mutations were absent. The pathology examination revealed abundant lymphocytic infiltration both inside and around the tumor. These lymphocytes were positive for CD3 but not for CD20 (Figure 5A, 5B, respectively), confirming that they were T lymphocytes. T lymphocytes were counted in 3 fields inside the tumor. The CD8-positive TILs (Figure 5C) and the CD4-positive TILs (Figure 5D) were similar in number, indicating approximately similar proportions of cytotoxic T cells and helper T cells in the tumor. His general condition was good, with no apparent organ dysfunctions. The 9LA regimen [6,7] (ie, ICIs (Nivolumab + Ipilimumab) and cytotoxic chemotherapeutic agents (carboplatin + paclitaxel) was started in September 20XX on admission. Except for slight pain in both feet, he remained asymptomatic after the initiation of this therapy. After the first course, the mass was decreased to 5.9 cm in maximum diameter (Figure 2B). He continued to receive outpatient immunotherapy after the second course. After the seventh course, the mass shrank to 3.8 cm (Figure 2C). Before the eighth course, in February 20XX, he developed grade 1 diarrhea and ground-glass opacity. Therefore, the eighth course was suspended for 3 weeks, and immunotherapy was resumed with nivolumab alone. The immunotherapy after the eighth course was carried out with no adverse events. The patient experienced disease progression after completing the 24th cycle of immunotherapy in October 20XX+1. Thereafter, he continued treatment with chemotherapy until his death in November 20XX+2.

Discussion

This NSCLC case featured PD-L1–negative status with abundant tumor-infiltrating lymphocytes (TILs) in the tumor microenvironment. Following treatment with the CheckMate 9LA regimen, a partial response was observed. This case is consistent with the findings of the CheckMate 9LA trial subgroup analysis, which demonstrated improved overall survival (OS) in several subgroups, including male patients, Asians, those with a baseline ECOG score of 0, current or former smokers, and patients with PD-L1 expression <1% [8]. This patient was 80 years old. Although the CheckMate 9LA trial did not demonstrate a statistically significant survival benefit in the subgroup of patients aged 75 years or older, this was likely due to the limited sample size in this cohort, as noted by the investigators.

ICIs exert antitumor effects by restoring T cell activity through blockade of immune-inhibitory pathways such as PD-1 and CTLA-4 [9]. Despite this mechanism of action, a favorable response was achieved even in the absence of PD-L1 expression. Current treatment guidelines support the use of immune-checkpoint inhibitors (ICIs) in patients with stage IV squamous cell carcinoma irrespective of PD-L1 expression, as long as the patient’s performance status permits [10,11]. Even in cases with less than 1% PD-L1 expression, the use of ICIs is recommended.

Tumor-infiltrating lymphocytes (TILs) are immune cells, primarily T cells, that have migrated into the tumor microenvironment. Their presence is often considered a marker of an ongoing immune response against the tumor, and high levels of TILs – particularly CD8-positive cytotoxic T cells – have been associated with improved outcomes in various cancers. Although there is no standardized guideline for the routine assessment of TILs in NSCLC, several studies have explored the prognostic and predictive significance of CD8+ T cell infiltration [12]. The International Immuno-Oncology Biomarker Working Group has proposed practical recommendations for TIL evaluation in solid tumors, though primarily in breast cancer and melanoma [13].

A retrospective analysis has indicated a positive correlation between PD-L1 expression and CD8-positive TIL abundance in non-small cell lung cancer [14]. In contrast, our case was notable for exhibiting dense infiltration of CD3-positive TILs despite being PD-L1–negative. Given the balanced infiltration of CD4+ and CD8+ T cells, CD3 staining in this context may still reflect a robust T cell-mediated immune response. One possible explanation for abundant infiltration is that the tumor microenvironment was immunologically active due to factors such as high neoantigen load or local inflammatory signals, which may have facilitated T cell recruitment. However, these infiltrating T cells may have been functionally suppressed through alternative immune-checkpoint pathways – including LAG-3, TIM-3, and TIGIT [15] – which are often upregulated in exhausted CD8-positive T cells. These molecules could permit T cell accumulation within the tumor while maintaining their dysfunctional state. The observed clinical response to combined PD-1 and CTLA-4 blockade may thus reflect partial restoration of antitumor immunity via reversal of non–PD-L1-mediated T cell exhaustion. This highlights a potential discordance between PD-L1 expression and tumor-infiltrating lymphocyte (TIL) density, suggesting that these parameters may be distinct aspects of the tumor immune microenvironment. Among them, TILs, particularly a CD3+/PD-L1− immune profile as observed in this case, may serve as an exploratory biomarker predicting response to anti-PD-1 and anti-CTLA-4 therapies, although they have not yet been established as a definitive biomarker comparable to PD-L1.

One of the limitations of this case report is that our patient received cytotoxic chemotherapy in addition to ICIs. Sakai et al reported that 7 of 14 patients showed increased PD-L1 expression on re-biopsy following chemotherapy [16]. In our case, we did not perform a repeat biopsy after treatment, and therefore we cannot completely rule out the possibility that the response to immunotherapy may have been influenced by chemotherapy-induced upregulation of PD-L1. Another limitation is the possibility that spatial heterogeneity within the tumor contributed to underestimation or inaccurate evaluation of PD-L1 expression [17].

Conclusions

This discordance between TPS and TIL density may suggest that immune suppression in the tumor microenvironment can occur through alternative pathways beyond PD-L1, such as LAG-3, TIM-3, or TIGIT, indicating the complexity and heterogeneity of tumor immune interactions. This case highlights the need for a more comprehensive understanding of the tumor immune microenvironment, especially in PD-L1–negative non-small cell lung cancer. In PD-L1–negative NSCLC, evaluation of TILs may aid in identifying patients likely to benefit from combined checkpoint blockade. Although a causal relationship cannot be inferred from a single case, the favorable clinical response observed alongside this immune profile suggests that the potential role of CD3-positive TILs warrants further investigation in larger studies.