Article Text
PDF
Abstract
Background In non-small cell lung cancer (NSCLC), chemoradiotherapy (CRT) yields pathological complete response (pCR) rates of approximately 30%. We investigated using ipilimumab plus nivolumab (IPI-NIVO) with neoadjuvant CRT in resectable, and borderline resectable NSCLC.
Methods This single-arm, phase-II trial enrolled operable T3-4N0–2 patients with NSCLC without oncogenic drivers. Primary study endpoints were safety, major pathological response (MPR) and pCR. Treatment encompassed platinum-doublet concurrent CRT, IPI 1 mg/kg intravenous and NIVO 360 mg intravenous on day-1, followed by chemotherapy plus NIVO 360 mg 3 weeks later. Thoracic radiotherapy was 50 or 60 Gy, in once-daily doses of 2 Gy. Resections were 6 weeks post-radiotherapy.
Results In a total of 30 patients in the intention-to-treat (ITT) population, grades 3–4 treatment-related adverse events (TRAEs) occurred in 70%, one TRAE grade 5 late-onset pneumonitis on day 96 post-surgery (1/30, 3.3%) occurred, and one non-TRAE COVID-19 death (1/30, 3.3%). pCR and MPR were achieved in 50% (15/30) and 63% (19/30) of the ITT; and in 58% (15/26) and 73% (19/26) of the 26 patients who underwent surgery, respectively. Postoperative melanoma was seen in one non-pCR patient. The R0 rate was 100% (26/26), and no patient failed surgery due to TRAEs. In peripheral blood, proliferative CD8+ T cells were increased, while proliferative regulatory T cells (Tregs) were not. On-treatment, pCR-positives had higher CD8+CD39+ T cells and lower HLA-DR+ Tregs.
Conclusions Neoadjuvant IPI-NIVO-CRT in T3-4N0–2 NSCLC showed acceptable safety with pCR and MPR in 58% and 73% of operated patients, respectively. No patient failed surgery due to TRAEs.
Trial registration number NCT04245514.
- Nivolumab
- Ipilimumab
- Chemotherapy
- Radiotherapy
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Patients with locally advanced borderline resectable non-small cell lung cancer (NSCLC), treated with chemoradiotherapy followed by surgery, experience significant local and distant disease relapse.
WHAT THIS STUDY ADDS
INCREASE added ipilimumab and nivolumab to standard chemoradiotherapy and surgery in borderline resectable NSCLC, achieving ca 60% pathological complete response (pCR) compared with the historical 30%, with no surgery failures due to treatment-related adverse events.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
These findings suggest that adding dual immunotherapy could enhance the efficacy of current standard care for locally advanced borderline resectable NSCLC. Further research is needed to determine the optimal combination of treatments for maximum efficacy with minimal toxicity.
Introduction
Non-small cell lung cancer (NSCLC) accounts for the highest incidence of cancer-related death worldwide.1 A subgroup of patients with NSCLC with large tumors, or tumors invading the chest wall or mediastinum (T3-T4), either with or without nodal involvement (N0-N2), can undergo surgery either at initial presentation, or following induction therapy. In selected patients, guidelines recommend resection after induction chemoradiotherapy (CRT) in order to maximize local and distant control rates.2 3 The use of multimodal strategies involving platinum-doublet chemotherapy combined with radiotherapy has improved long-term clinical outcomes, with acceptable toxicity.2 4–7 However, overall survival (OS) is impaired by distant disease relapses which underlines the need for improved systemic tumor control.8–10
Neoadjuvant immunotherapy can improve local and distant control rates in patients presenting with resectable NSCLC. A phase-II trial compared neoadjuvant ipilimumab (anti-cytotoxic T-lymphocyte associated protein 4 (CTLA-4), IPI) plus nivolumab (anti-programmed cell death protein-1 (PD-1), NIVO) to neoadjuvant NIVO alone, in patients with stages I-IIIA NSCLC, and reported pathological complete response rates (pCR) of 29% with IPI-NIVO and 9% with NIVO alone in the intention-to-treat (ITT) population, and in the resected patient population: 38% with IPI-NIVO and 10% with NIVO alone.11 Furthermore, synergy may arise from combining immunotherapy with radiotherapy.12 These findings suggested that antitumor immune activity could be enhanced through synergy between radiotherapy and anti-CTLA-4 therapy.
We postulated that adding IPI-NIVO to CRT would further enhance pCR rates and antitumor immune responses. Therefore, the aim of this study was to investigate the addition of IPI-NIVO to standard induction CRT in patients with resectable and borderline resectable T3-4N0-2 stage IIB-III NSCLC tumors to determine safety and pathological response.
Materials and methods
Study population and design
Participants and screening
Between January 2020 and July 2022, this single-center, single-arm prospective phase-II study (INCREASE trial) enrolled fit patients with pathology-proven cT3-4N0-2M0 NSCLC according to the tumor, node, metastases eighth edition.13 Patients with actionable genomic alterations like activating EGFR or BRAF mutations or ALK or ROS1 gene rearrangements were excluded. A list of tested genomic alterations is provided in online supplemental table 1. Staging was performed according to European Society for Medical Oncology (ESMO) guidelines, including a tumor biopsy, pulmonary function tests, an 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET)/CT scan, mediastinal staging using EBUS and brain imaging.2 No patient underwent a mediastinoscopy. At two time points, a decision on resectability was taken within the multidisciplinary team (MDT) by at least two experienced surgeons, namely: during screening and after finishing CRT. Criteria for resectability included cT3-4N0-2, excluding those T3-4 based on satellite lung lesions, those based on ingrowth in the heart, aorta, trachea, and esophagus, and excluding multistation N2. The criteria for operability were determined based on the preoperative risk assessment tests recommended by ESMO.2 The clinical trial protocol, including the full list of inclusion and exclusion criteria, has been published.14
Supplemental material
Procedures overview
A detailed description of the pre-surgical and post-surgical diagnostic and treatment procedures are given in the online supplemental data (chapter 1). In brief, patients received two cycles of chemotherapy concurrent with thoracic radiotherapy to a dose of 50 Gy, delivered in once-daily fractions of 2 Gy. For cases judged to be “borderline resectable” by surgeons, the total radiation dose delivered was 60 Gy.6 On day-1 of chemotherapy, one cycle of IPI (1 mg/kg intravenous) plus NIVO (360 mg flat dose intravenous) (IPI/NIVO) was administered. A second cycle of NIVO (360 mg flat dose intravenous) plus chemotherapy was administered 3 weeks later.
Clinical outcomes
The co-primary study objectives were to assess the (1) safety and (2) pathological response using major pathological response (MPR) and pCR in the ITT population and in the evaluable patients population, that is, those who completed induction therapy, including at least one cycle of immunotherapy, followed by surgery.15 Secondary objectives were to assess disease-free survival (DFS), defined as the time from resection to recurrence of tumor or death, and OS, defined as the time from study inclusion to death. Exploratory objectives included assessing the immune competence of tumor-draining lymph nodes, and changes in peripheral blood immune subset distributions and phenotypes. The definitions used for primary, secondary and exploratory endpoints were as reported in our previously published trial protocol and are also described in the online supplemental data (chapter 2).14
Tumor tissue, peripheral blood mononuclear cell analyses
A detailed description of the procedures regarding the pathological evaluation of tumor response, programmed death-ligand 1 (PD-L1), tumor genomic analysis, peripheral blood mononuclear cell (PBMC) analyses, and the depiction of the gating strategy and the specific markers used, including those for the regulatory T cells (Treg) population, is included in the (online supplemental data chapter 3, figure 1, tables 1,2).
Statistical analysis
Per the study protocol, the co-primary aims were safety and assessing pathological response, and both pCR and MPR were considered as endpoints. However, as pCR has been more consistently reported in published papers for “historical” comparison, and as pCR is less prone to differing assessment criteria and interobserver variability, we decided to choose pCR to estimate our sample size and report as main efficacy endpoint.6 8 9
As compared with the previously reported rates of pCR following CRT averaging around 30%, this study aimed to test whether the pCR rate would double to 60%, using a Z-test for a single proportion with a two-sided significance level of 5%. A total of 26 evaluable patients were needed to reach 90% power to reject the null hypothesis if pCR after IPI-NIVO-CRT was 60%. To account for a drop-out rate of 10%, the planned accrual was 29 patients.
Descriptive statistics (proportions with 95% CIs) were used to summarize the endpoints, using frequencies and percentages for categorical variables and as median and range for continuous variables. Differences between categorical and continuous variables were assessed by t-test or χ2 test, when appropriate. For translational analyses with multiple measurements, two-way analysis of variance was used to compare pCR and non-pCR patients per time point; corrections for multiple comparisons were performed according to either Tukey or Šidàk, when appropriate. Detailed results with p values are provided in the online supplemental data (chapter 6).
Results
Patients disposition and characteristics
30 patients commenced induction therapy, of which 26 (87%) underwent post-induction surgery (see figure 1). Four patients did not undergo surgery, which in one case was due to death from respiratory failure due to a COVID-19 infection during induction therapy. Three patients failed to undergo surgery because of disease progression (N=1), new lung nodules that were later determined to be pseudo-progression (N=1), and an MDT recommendation to opt for adjuvant durvalumab following a post-induction metabolic complete response in a patient who would have had to undergo resection of three vertebrae (N=1). Of 26 patients who underwent surgery, 1 was excluded from further analysis of clinical outcomes as the pathology specimen revealed metastasis from undifferentiated melanoma.
The patient disposition in the INCREASE trial. (*) This patient achieved a metabolic complete response after induction chemo-immuno-radiotherapy and the MDT recommended to omit surgery that would have involved a most likely futile resection of three vertebrae. MDT, multidisciplinary tumor board; NSCLC, non-small cell lung cancer.
Patient characteristics are summarized in table 1. The median patient age was 64 years (range 43–73), most were current or former smokers (97%) and a majority presented with non-squamous histology (72%). Tumor PD-L1 expression was 50% or more in 63% of patients. A comparison of baseline characteristics of patients with and without pCR is shown in online supplemental table 3.
Baseline patient characteristics
Adverse events and surgical outcomes
Adverse events were scored in all 30 patients who commenced induction therapy. All patients experienced treatment-emergent adverse events (TEAEs), with 25 (83%) patients experiencing grade 3 or higher toxicity (online supplemental table 4). Table 2 summarizes treatment-related adverse events (TRAEs) recorded in at least three (10%) patients (a full list is provided in online supplemental table 5). One or more grade 3 immune-related adverse events were recorded in seven (23%) patients, leading to discontinuation of NIVO during the second cycle in 10% of patients. Two patients died during the study: one due to COVID-19 and another due to pneumonitis-induced respiratory failure 96 days after surgery. The latter developed rapidly progressive interstitial lung disease approximately 100 days after the last fraction of radiotherapy, for which no infectious cause was identified. The fatal event was scored as a late treatment-related immune-related adverse event.
Treatment-related adverse events
A total of 26 out of 30 patients underwent surgery, resulting in an 87% resection rate in the ITT population. The median time from the last radiotherapy fraction to surgical resection was 42 days (range 35–63 days). TRAE’s did not lead to failures to undergo surgery. A total of 25 lobectomies (96%) and 1 pneumonectomy (4%) were performed, all with a systematic nodal dissection. A pathological complete (R0) resection was achieved in all patients. Surgical complications occurred in 16/25 patients, scored as Clavien-Dindo grade 2 complications in 12 patients, grade 3a complications in 3 patients, and grade 3b complications in 2 patients (20% grade 3a-b complications). No 30-day or 90-day mortality was observed. A pneumo-pleural fistula developed in one patient, which resolved spontaneously on antibiotics.
Pathological responses
Of the 26 patients who underwent a resection, 7 patients, including the patient with pulmonary melanoma, had more than 10% residual viable tumor cells. From the remaining patients, 15 (58%; 95% CI: 37% to 77%) achieved a pCR, and the other four had an MPR (73%; 95% CI: 52% to 88%). For the 25 operated patients with NSCLC (excluding the single patient with melanoma), the pCR and MPR rates were 60% (95% CI: 39% to 79%) and 76% (95% CI: 55% to 91%). For the ITT, these were 50% (95% CI: 31% to 69%) and 63% (95% CI: 44% to 80%), respectively (figure 2).
(A) The waterfall plot for pathological response (top row), shows that 15 (60%) out of 25 operated patients with NSCLC (excluding the patient with melanoma) achieved a pCR (dark blue bars), another 4/25 (16%) achieved an MPR (light blue bars), the remaining 6 (24%) patients had more than 10% residual viable tumor cells left (orange bars). In operated non-squamous cell lung cancers, genomic alterations associated with resistance to immunotherapy (EGFR, ERBB2, STK11 and NRAS) were seen in eight patients, all of which had no pCR. An STK11 mutation was seen in the only patient who developed PD on induction therapy (ie, patient 15). TP53 was seen in 81% (13/16) of operated patients where genomic analysis was performed, while only one tested patient had a KRAS mutation. Only 28% (7/25) of operated patients with NSCLC had squamous cell histology (dark green). All, except for one patient, were current or former smokers. The only never-smoker (patient 03) had an EGFR exon 21 (L858R) mutation, which was found post-surgery. High tumor PD-L1 expression rates were observed in 56% (14/25) of patients with resected NSCLC (bottom row). (B) Radiological tumor responses to the induction therapy, that is, change in the sum of longest diameters according to RECIST V.1.1, are shown in the patients with NSCLC. Partial response was seen in 11 patients and stable disease in 16 patients. In patient 32, pseudo-progression with new lung nodules were seen, while the primary tumor shrunk. In patient 15, PD was seen due to the appearance of new pleural lesions on induction therapy, while the treated baseline tumor lesions shrunk. Patient 19 did not complete induction therapy and could therefore not be evaluated. Patients with pCR (dark blue), MPR (light blue) and no MPR (orange) can be seen among the patients with radiological partial response and stable disease. KGA, key genomic alterations; MPR, major pathological response; NSCLC, non-small cell lung cancer; pCR, pathological complete response; PD, progressive disease; PD-L1, programmed death ligand-1; VTC, viable tumor cell.
Baseline tumor PD-L1 expression of 50% or above was seen in 63% of patients, however, the presence of high PD-L1 expression was not statistically different among patient groups based on the presence of a pCR or MPR. Online supplemental table 6 shows the full list of PD-L1 results.
Six out of the eight patients with non-squamous NSCLC who underwent surgery but did not achieve pCR were found to have genomic alterations known to be associated with resistance to immunotherapy (figure 2A).16–19 The single patient who developed disease progression on induction therapy had an STK11 mutation. Detailed results from the molecular analysis is shown in online supplemental table 7.
Radiological and 18F-FDG responses
Figure 2B summarizes the depth of radiological responses on CT scan to induction therapy, that is, change in the sum of longest diameters according to Response Evaluation Criteria in Solid Tumors (RECIST) V.1.1, in the ITT NSCLC population. No patient achieved a radiological complete response. Partial responses (PR) were seen in 11 (37%) patients, and stable disease in 16 (53%) patients (SD). pCR was not different between patients with PR and those without (p=0.096). Two patients developed new lesions on induction therapy while the treated baseline tumor lesion shrunk. Of these, one patient developed pleural lesions that were scored as progressive disease (PD), and one patient developed bilateral new pulmonary lesions during therapy, which were initially scored as PD, but subsequently regressed spontaneously, and were scored as pseudo-progression.
Figure 3 illustrates the radiological and metabolic responses observed in an adult patient with an adenocarcinoma in the left upper lobe invading adjacent thoracic vertebra and extrathoracic tissues. The post-surgical pathology specimen revealed a pCR. The patient was discharged 11 days after surgery, and remains free of disease 16 months after resection, without a need for pain medication.
The FDG PET/CT images of an adult patient (23), before (top row) and after (bottom row) induction chemo-immuno-radiotherapy are shown, using fused PET/CT images (A,E) and CT-attenuation corrected PET images of axial (B,F), sagittal (C,G) and coronal view (D,H) reconstructions. This patient had a tumor in the left upper lobe with, at baseline, ingrowth in the thoracic wall and adjacent structures such as the musculature and left ribs (red arrows). The tumor had a necrotic center (N), and there was no apparent mediastinal lymph node involvement. CTAC, CT-attenuation correction; FDG, fluorodeoxyglucose; PET, positron emission tomography; RECIST, Response Evaluation Criteria in Solid Tumors.
Survival outcomes
Per August 2023, at a median follow-up time of 25.8 months (range 1.0–42.3), in 8 out of 30 patients, disease recurrence was seen, and 7 patients died. These initial survival outcomes, including their respective events, are shown in online supplemental figure 6 and table 8.
Immunological responses
As resected tumor tissues were mostly necrotic, these were found to be unsuitable for substantive analysis.
Peripheral blood immune monitoring
Figure 4A displays the baseline immune checkpoint expression as a percentage of the CD8+T cell and Treg populations. Alongside this, the mean fluorescence intensity is presented, which aligns well with the percentages, highlighting relative differences in immune checkpoint expressions. At baseline, CD8+T cells predominantly expressed PD-1 and T cell immunoreceptor with immunoglobulin and ITIM domain (TIGIT) over Lymphocyte Activation Gene 3 (LAG3) and CTLA-4. In Tregs, LAG3 had the least expression, followed by PD-1, then TIGIT, with CTLA-4 being the highest. Among patients without pCR, most immune checkpoint expressions were similar or higher (non-significant) than in those with pCR, both in CD8+T cells as well as Tregs. Notably, only CTLA-4 expression in Tregs was significantly higher in patients without pCR.
Peripheral blood T-cell subset analysis and its correlation with pathological response. The tests used and the exact p values are shown in the (online supplemental data chapter 6 figures 2–5), here, asterisks indicate significant differences (p<0.05). (A) Immune checkpoint expression in CD8+T cells (red, first two panels) and regulatory T cells (blue, last two panels) from 13 patients at baseline. Data is shown as a “percentage of the T cell population”, which indicates the proportion of cells expressing a marker within the total CD8+T cell or Treg population, as well as the “mean fluorescence intensity” for each T-cell subset. Solid symbols denote patients with pCR (N=7) and hollow symbols denote those without pCR (N=6). Medians and ranges are displayed for each immune checkpoint. (B) Longitudinal expression of LAG3, TIGIT, and CTLA-4 in CD8+T cells (red) and Tregs (blue) for 11 patients at baseline, surgery, and 12 weeks post-surgery as a percentage of the total population of T-cell subsets (either CD8 or Treg). Solid symbols represent patients with pCR (N=6) and hollow symbols represent those without pCR (N=5). (C–D) Markers for proliferation (Ki67), activation (HLA-DR), and tumor association (CD39) in CD8+T cells (figure 4C) and Tregs (figure 4D) are shown for 11 patients at baseline, surgery, and 12 weeks post-surgery as a percentage of the total population of T-cell subsets (either CD8 or Treg). Solid symbols denote patients with pCR (N=6) and hollow symbols represent those without pCR (N=5). CTLA-4, cytotoxic T-lymphocyte associated protein 4; LAG3, Lymphocyte Activation Gene 3; pCR, pathological complete response; TIGIT, T cell immunoreceptor with immunoglobulin and ITIM domain; Treg, regulatory T cells.
Reliable post-induction PD-1 measurements were hindered by interference from therapeutic anti-PD-1 with the Fluorescence-activated cell sorting (FACS) antibody. This was not a problem for intracellularly measured CTLA-4. Figure 4B presents longitudinal immune checkpoint changes. LAG3 was consistently low, with a slight drop in CD8+T cells of non-pCR patients from baseline to surgery. TIGIT was stable initially but increased post-surgery, significantly so in pCR Tregs. While CTLA-4 expression in CD8+T cells was low at baseline and stayed low, its expression in Tregs was notably higher and showed an increasing trend over time. Notably, CTLA-4 expression was significantly higher in patients without pCR compared with those with pCR, both at baseline and overall (p=0.0197).
Figure 4C–D displays Ki67+, HLA-DR+, and CD39+expression on CD8+T cells and Tregs at baseline, surgery, and 12 weeks post-surgery. Ki67 indicates cell proliferation, HLA-DR signals T-cell activation, and CD39 suggests tumor-reactivity.20 Post-induction, CD8+T cells showed increased Ki67 and HLA-DR levels, both reverting to baseline post-surgery. CD39+CD8+ T cells rose post-induction and remained elevated 12 weeks post-surgery, especially in patients with pCR (p=0.0476). Ki67+Treg levels remained stable during induction therapy and post-surgery. However, upregulation of HLA-DR was observed in patients who failed to achieve a pCR, and CD39+Treg frequencies showed a trend towards an increase throughout treatment and follow-up (non-significant).
Further testing details are in the supplementary data, namely online supplemental figures 2–5 and the PBMC test comparison tables.
Discussion
To the best of our knowledge, the INCREASE study is the first to investigate the safety and efficacy of adding dual immunotherapy, using a PD-1 and a CTLA-4 blocker, to high-dose concurrent CRT in the neoadjuvant setting for locally advanced NSCLC. Our main findings were a pCR and MPR in 50% (15/30) and 63% (19/30) of the ITT population, and in 58% (15/26) and 73% (19/26) of patients who underwent surgery, respectively. The addition of dual immunotherapy to CRT substantially increased pCR rates above historical rates after CRT, which vary from 21% to 45%.5 6 8 A summary of pathological responses to various neoadjuvant regimens are provided in table 3.
Comparison of pathological outcomes
The rationale and results of our study closely align with those of the recently published platform NEOSTAR study, which reported increased pathological responses using the combination of NIVO and IPI with chemotherapy in resectable stage IB-IIIA NSCLC.21 The study successfully met its primary endpoint of MPR, with 32.1% (7/22, 80% 95% CI: 18.7% to 43.1%) in the NIVO+CT arm and 50% (11/22, 80% 95% CI: 34.6% to 61.1%) in the IPI+NIVO+CT arm.
While MPR or pCR are not a survival endpoint, following neoadjuvant therapy, they have been found to potentially predict DFS and OS, regardless of the neoadjuvant treatment modality. Pataer et al showed that in patients with resectable NSCLC, MPR following neoadjuvant chemotherapy was associated with a longer OS.22 Analysis of our previous institutional outcomes for sulcus superior tumors treated with CRT and surgery identified pCR as predictive for improved 5-year OS, with an HR of 0.27 (95% CI: 0.15 to 0.50; p<0·001).6 In the phase-III CheckMate-816 study, investigating neoadjuvant chemotherapy with or without NIVO in resectable NSCLC, the residual percentages of vital tumor cells also predicted for event-free survival at 2 years in the chemo-NIVO arm (area under the curve (AUC)=0.74).23 Of note, although the high PD-L1 expression levels that were observed in 63% of patients in this study might be expected to influence pathological responses, pCR were observed in tumors expressing high, low, and absent PD-L1 expression levels. This is also in line with the observations in the IPI-NIVO-CT arm of the NEOSTAR platform study, but also CheckMate-9LA and CheckMate-227, where the efficacy of IPI-NIVO-CT was irrespective of tumor PD-L1 expression levels.21 24 25
Despite excluding patients with an identified targetable oncogenic driver before study inclusion, surgical specimens revealed the presence of such drivers. One patient with an EGFR exon 21 (L858R) mutation had 30% residual viable tumor cells in the resected tumor, and another patient with an ERBB2 mutation had 70% residual viable tumor cells post-induction therapy. In addition, four patients with STK-11 mutations were identified, one of whom had a pCR and one developed disease progression post-induction therapy. These findings suggest that genomic features associated with immunotherapy resistance may adversely influence the efficacy of the IPI-NIVO-CRT combination. However, caution is warranted in view of the limited number of patients.
As reported previously for neoadjuvant NSCLC studies, radiological responses observed to this induction therapy did not correlate with pCR.26 27 Indeed, INCREASE patients with a pCR showed a best radiological response of PR in 53%, and SD in 47%. These findings highlight a need for improving response evaluation to neoadjuvant therapies.
In the ITT population, 83% suffered grade ≥3 TEAEs and 73% encountered TRAEs. However, these toxicities did not prevent surgery, and toxicity rates aligned with CRT studies like the PROCLAIM study, where 64% and 76.8% grade 3–4 toxicities were reported.28 The current study identified immune-related events, with 23% being grade ≥3, including one fatal pneumonitis. A recent study combining CRT with IPI and NIVO reported a 63% pulmonary toxicity rate.29 Our lower pneumonitis rate might be the result of resecting irradiated lung tissues and using lower radiation doses (50 Gy in 86% of patients). The influence of unresected irradiated lung tissue on high-grade pulmonary toxicity needs further exploration.30
The remarkable improvement in pathological tumor response observed with the IPI/NIVO and CRT combination might be a result of enhanced immunogenic cell death, caused by the proliferation and activation of effector T cells and the suppression of Tregs. Indeed, PBMC analysis revealed a noticeable post-induction surge in proliferative and activated CD8+effector T cells, as evidenced by increased Ki67+and HLA-DR+populations. However, this surge returned to baseline levels post-surgery. Intriguingly, CD39, a marker linked to tumor-reactivity, demonstrated a post-induction rise that was sustained for up to 12 weeks post-surgery, suggesting the continuing presence of tumor-reactive effector T cells. Meanwhile, suppressive Tregs did not proliferate, and interestingly, their activation was notably less in patients achieving a pCR.
Baseline data revealed that CD8+T cells expressed CTLA-4 less than other immune checkpoints, whereas Tregs had a markedly high CTLA-4 expression. Also, CTLA-4 expression in Tregs was even higher in patients without a pCR compared with those achieving pCR. When observing longitudinal changes, CTLA-4’s expression pattern stood out. While its levels in CD8+T cells remained consistently low, Tregs exhibited a pronounced and increasing expression over time, suggesting that IPI’s blockade of CTLA-4 could be of considerable therapeutic importance in reducing the immune suppressive effects of Tregs.
These findings align well with the results of the NEOSTAR study, particularly the IPI-NIVO-chemotherapy arm, showing increased CD3+CD8+ tumor-infiltrating T cells and heightened antigen-activated T cells post-therapy.21 However, NEOSTAR also showed some interesting leads into other cell lines such as B-cell abundance and its inverse correlation with viable tumor cells, suggesting an association with immunotherapeutic response. Additionally, CXCL9+tumor-associated macrophages exhibited a functional role in response to immune checkpoint therapy, indicating enhanced immune activation, effector memory, cytotoxic function, and reduced immunosuppressive cell subsets in treated tumors.
Some limitations of this exploratory single-center, single-arm study are acknowledged. Despite a significant increase in pathological response, namely pCR rates, it is unclear whether pCR alone could function as a reliable surrogate endpoint, even though data from large neoadjuvant studies is suggestive of this.23 31 32
A substantial number of TRAEs was observed, which, although manageable, suggests that critical assessment of patient selection, drug dosage and therapy de-escalation all deserve further study. Other trials are currently investigating the immune-modulatory effects of different fractionation regimens of radiotherapy to enhance the effects of neoadjuvant PD-L1 blockade after neoadjuvant chemotherapy in patients with resectable stage III (N2) NSCLC.
The expected outcome of this trial was a 60% pCR rate, this was not achieved in the ITT and resected populations, the operated NSCLC population did reach this rate, with 15 out of 25 patients achieving pCR (60%). However, the initial sample size calculation was powered at 90%, but when considering an 80% power, both the ITT and resected populations meet the criteria. It is important to emphasize that this was an exploratory single-arm phase 2 study designed to assess the impact of adding dual immunotherapy to CRT rather than strictly meeting the 60% pCR benchmark.
The postoperative resection specimens, collected on average 6 weeks post-radiotherapy, were predominantly necrotic. This timing likely led to the loss of tumor immune dynamics and profiles typical of dual immunotherapy with CRT within the resection specimens. Consequently, the remaining tissue primarily consisted of mainly necrotic tumor beds, unsuitable for meaningful translational analysis.
This regimen is unlikely to be adopted in routine clinical practice due to its complexity and the necessity for it to be conducted in highly specialized centers that are proficient in complex surgeries post-CRT and managing severe toxicities. Despite this, the regimen fills a critical knowledge gap and provides valuable insights into the multimodality treatment landscape. There is an ongoing need to enhance treatment outcomes while minimizing toxicity and patient burden. The ultimate goal within the scientific community is to refine the dose, sequence, and frequency of these multimodality treatment components, including various immunotherapies combined with chemotherapy, radiotherapy, and surgery, to achieve optimal local and distant control with the least amount of adverse events.
Conclusion
This study revealed that the use of neoadjuvant dual immunotherapy with CRT in T3-4N0–2 NSCLC resulted in enhanced pCR and MPR with acceptable surgical morbidity. Patients with pCR demonstrated higher proliferation and activation rates of CD8+ T cells and lower rates in Tregs, these observations warrant further investigation.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
Ethical approval was obtained for this study from the institutional review board at Amsterdam UMC, location VUmc (Amsterdam, The Netherlands), with study number 2019.710. Participants gave informed consent to participate in the study before taking part.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
X @ezgi_ulas
Contributors IB: Conceptualization; Project administration; Funding acquisition; Data collection, curation, analysis; Supervision; Validation; Writing—original draft; Writing—review and editing. CD: Conceptualization; Project administration; Funding acquisition; Data collection, curation, analysis; Supervision; Validation; Writing—original draft; Writing—review and editing. FLS: Data collection, curation, analysis; Validation; Writing—original draft; Writing—review and editing. JV: Data collection, analysis; Writing—review and editing. DJH: Data collection, analysis; Writing—review and editing. SMSH: Data collection, analysis; Writing—review and editing. AV: Data collection, analysis; Writing—review and editing. IH: Data collection, analysis; Writing—review and editing. EBU: Data collection, analysis; Writing—review and editing. JB: Data collection, analysis; Writing—review and editing. PvdV: Data collection, analysis; Writing—review and editing. NB: Data collection, analysis; Writing—review and editing. LJM: Data collection, analysis; Writing—review and editing. DEO-L: Data collection, analysis; Writing—review and editing. FvM: Data collection, curation, analysis; Validation; Writing—original draft; Writing—review and editing. MFF: Data collection, curation, analysis; Validation; Writing—original draft; Writing—review and editing. TDdG: Data collection, curation, analysis; Validation; Writing—original draft; Writing—review and editing. TR: Data collection, curation, analysis; Validation; Writing—original draft; Writing—review and editing. SS: Conceptualization; Project administration; Funding acquisition; Data collection, curation, analysis; Supervision; Validation; Writing—original draft; Writing—review and editing. Accountable for all aspects of the work: IB (corresponding author). All authors approved the final version of the manuscript. IB is the guarantor for the overall content.
Funding This study was funded by Bristol-Myers Squibb (BMS) (CA209-7FP). M F Fransen and J Bakker were funded by the Dutch Cancer Society (KWF 14102).
Competing interests The authors have disclosed the following conflicts of interest: DEO-L received grants from Janssen and Curium, honoraria from EANM, EAU, ESMO, and Curium, travel support from EANM, ESMO, EAU, and Bayer, and serves on a Bayer advisory board. CD received institutional fees for advisory roles with BMS, AstraZeneca, and MSD. SMSH has research contracts with several pharmaceutical companies and received honoraria from Janssen. IB’s institution received support from BMS, AstraZeneca, and Boehringer Ingelheim, and he received honoraria from AstraZeneca, MSD, and BMS. FLS received a research grant from ViewRay and honoraria from AstraZeneca. SS’s institution received grants from Varian, ViewRay, and AstraZeneca; he received consulting fees and honoraria from AstraZeneca and MSD, and serves on advisory boards for AstraZeneca and MSD. TDdG’s institution received grants from Idera Pharmaceuticals, consulting fees from LAVA Therapeutics and Mendus, and she holds stock in LAVA Therapeutics. All other authors have declared no conflicts of interest.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.