A prospective study of immune responses in patients with lung metastases treated with stereotactic body radiotherapy with or without concurrent systemic treatment

We prospectively evaluated the effects of stereotactic body radiotherapy (SBRT) on circulating immune cells. Patients with oligo-metastatic and oligo-progressive pulmonary lesions were treated with SBRT with (cSBRT) or without (SBRT group) concurrent systemic treatment (chemotherapy or immune checkpoint blockade) using different fractionation regimes. Immunoprofiling of peripheral blood cells was performed at baseline, during, at the end of SBRT, and at the first and second follow-ups. The study accrued 100 patients (80 with evaluable samples). The proportion of proliferating CD8+ T-cells significantly increased after treatment. This increase remained significant at follow-up in the SBRT group, but not in the cSBRT group and was not detected with doses of >10Gy per fraction indicating that lower doses are necessary to increase proliferating T-cells’ frequency. We detected no favorable impact of concurrent systemic treatment on systemic immune responses. The optimal timing of systemic treatment may be post-SBRT to leverage the immune-modulating effects of SBRT.


Introduction
Stereotactic body radiotherapy (SBRT) is emerging as an essential treatment modality in the oligometastatic setting [1][2][3][4][5] .SBRT can provide local tumor control through signi cant cytotoxic effects, due to the induction of the double-strand breaks, but can also have systemic effects on the anti-tumor immune response, as previously described 6,7 .Several studies have suggested that using SBRT with sub-ablative doses can lead to immunogenic cell death in the irradiated lesion, and induce immune-mediated abscopal effects outside the irradiation eld 8,9 .These effects remain poorly understood, and it is unclear whether the combination of SBRT with systemic treatments including immune checkpoint blockade (ICB) can lead to greater systemic responses because of potential antagonistic interactions.The question of optimal dose and fractionation and timing of immunotherapy or chemotherapy for the enhancement of immune modulation remains of great importance.
We have recently reported that SBRT can induce a signi cant decrease in the mean absolute counts of CD8 + cytotoxic T lymphocytes (CTLs) and CD4 + T-cells in early-stage non-small cell lung cancer (NSCLC) patients 10 .Nevertheless, the proportion of proliferating CD4 + and CD8 + T-cells among peripheral blood lymphocytes signi cantly increased after SBRT.Additionally, we have shown that the signi cant increases in the proportion of CD8 + and CD4 + proliferating T-cells compared to pre-treatment values were only detected in the NSCLC patients who received SBRT with 10Gy or less per fraction.
In this study, we evaluated the effects of SBRT on circulating immune cells when using different fractionation regimes, with or without systemic treatment (including ICB) in the oligometastatic and oligoprogressive disease settings in the lung.

Materials and Methods
Patients and study design.The prospective study was conducted in the Departments of Radiation Oncology of the University Medical Center Freiburg, Germany, and the Massachusetts General Hospital (MGH) and Harvard Medical School, Boston, USA, according to the Declaration of Helsinki.The study was registered in the German trials registry (DRKS 00011266).All patients gave written informed consent according to institutional and federal guidelines.The institutional ethics committees approved the study protocol (LAPIS trial, EK 38/16, Freiburg and MGH IRB Agreement #:2016D009860).
The prospective study included of two arms, depending on the localization of the tumors (lung or liver).The study enrolled patients with primary tumors or oligometastatic/oligoprogressive disease treated with SBRT with or without concurrent systemic treatments such as chemotherapy or immunotherapy (Supplementary Table 2).Each subgroup was analyzed separately, per protocol.We have previously reported the results in early-stage NSCLC 10 .Herein we present the results of the patients with metastatic pulmonary lesions treated with SBRT or hypofractionated radiotherapy with or without systemic treatment.We used immunopro ling of peripheral blood cells by longitudinal assessment at rst SBRT fraction (baseline), during and at the end of SBRT, and at rst (FU1) and second (FU2) follow-up (1½ and 4½ months after SBRT, respectively).
Treatment planning and treatment delivery.Patients were immobilized in a supine position with a customized vacuum cushion system and received a 4D/CT or a 4D/PET-CT.Patients with peripheral tumors not abutting the chest wall received 3 x 18.75Gy to the D50% such that 95% of the PTV received a minimum of 45 Gy (3 x 15Gy, 80% of the nominal dose) and a dose maximum between 110 and 120% 11 .Depending on the proximity to the central bronchial system and for tumors abutting or overlapping with the chest wall, a total dose of 50Gy in 5 fractions of 10Gy 12 or 60Gy in 8 fractions of 7.5Gy for central tumors 13 or 66 Gy in 12 fractions for ultra-central tumors was applied, as previously described 10 .One patient was treated with hypofractionated palliative radiotherapy with 3Gy per fraction up to a total dose of 39 Gy(major deviation).The dose prescription was chosen so that 95% of the PTV received at least the nominal fraction dose, and 99% of the PTV received a minimum of 90% of the nominal dose.The dose maximum within the PTV was chosen to be more than 110% but less than 120% of the prescribed dose.Response to treatment was assessed at the same time points according to the Response Evaluation Criteria in Solid Tumors (RECIST) using thoracic CT and 18F-FDG PET/CT, the latter being mandatory in case of suspected disease progression.Flow cytometric analysis.Blood samples were collected before treatment (baseline), 1 day after (during), at the end (end), at the 1st follow-up (FU1: six weeks after the end of SBRT), and the 2nd follow-up (FU2: 3 months after FU1).Peripheral blood mononuclear cells (PBMCs) were isolated and frozen until use.Not all samples were available from each patient at each time point.The reason for the missing samples was that there were either not collected or had insu cient cells for all analyses.
Statistical analysis.The null hypothesis for the primary endpoint was that the probability p of an increase in CD8 + counts at FU1 compared to baseline was p ≤ 50%, to be tested against the alternative that p > 50%.With 50 patients per subgroup, an exact one-sided binomial test at a 5% signi cance level had at least 80% power to reject the null hypothesis if p > 68.5% applied, and the null hypothesis would be rejected if at least 32 out of 50 patients experienced an increase (exact signi cance level 3.25%, STPLAN version 4.5).Intra-individual changes in blood biomarkers compared to baseline were examined using a mixed effects model for repeated measures, and multiple comparisons were corrected using the Benjamini, Krieger, and Yekutiely method to control a False Discovery Rate of 5% within variables over time.In a second step, we determined the standardised changes from baseline.Secondary endpoints included changes in other T-cell subsets at all time-points.
Overall (OS) and progression-free survival (PFS) were calculated from the start of SBRT and estimated according to the Kaplan-Meier method.For event-free patients, the observations for OS were censored at the date of the last contact and for PFS at the time of the last imaging.No patients died between baseline and end of treatment.To investigate the correlation between parameters signi cantly changed at FU1 or FU2 compared to baseline, PFS was calculated from FU1 or FU2 in landmark analyses in the patients still at risk (alive and without progression at FU1 and FU2), respectively.
All p-values except for the primary null hypothesis were two-sided and considered signi cant if below 5%.The statistical analysis used Prism (Prism V.8, GraphPad Software) and SPSS software (IBM, SPSS, v27).

Results
Patient characteristics.Between 2016-2022, the study accrued 63 patients with oligoprogressive or oligometastatic disease and received no concurrent systemic treatment within 3 months before SBRT (SBRT group) of whom 50 patients with 55 lesions were evaluable by ow cytometry (6 did not receive radiotherapy, 7 patients did not have su cient blood cells).The median time between the last treatment and SBRT was 25 months.Additionally, a second group which consisted of 37 patients who received concurrent systemic treatment (12 immunotherapy, 18 chemotherapy, or other systemic treatments) in combination with SBRT (cSBRT group) was included in the study.Of the 37 patients, 30 patients with 32 lesions were evaluable in this group (5 patients received no radiotherapy and 2 withdrew consent), of whom 12 received concurrent immunotherapy and 18 concurrent chemotherapy or other systemic treatments.
A total of 25 patients in the SBRT group and 18 patients in the cSBRT group had oligoprogressive disease (OPD) showing signs of progression in one metastatic lesion while having further measurable metastases without signs of progression that were not treated with other local treatments.Patients and treatment characteristics are shown in Table 1 and Supplementary Table 1.
Treatment outcomes.With a median follow-up of 30 months, the median OS was 30 months in the SBRT group and 53 months for cSBRT group, and the median PFS was 13 months in the SBRT group and 5 months for cSBRT group.Of note, patients with concurrent ICB treatments had a median PFS of 24 months.In the SBRT group, only one patient developed local progression and declined further treatments, 18 patients developed a regional progression treated with systemic therapy, radiotherapy (mostly SBRT), or resection, and 6 patients developed distant metastases.One patient developed a loco-regional recurrence, 1 patient developed a local recurrence in combination with distant metastases, and 1 patient developed a regional progression in combination with this metastasis.In the cSBRT group, 11 patients developed a regional progression, 7 patients developed distant metastases, and 1 patient had local and distant progression.All types of progression and subsequent treatments are summarized in Supplementary Table 2.
In the 25 patients in the SBRT group with oligoprogressive disease, 15 showed progression during followup while 9 had stable disease (one refused CT scans at follow-up).In the cSBRT group, of the 18 patients with oligoprogressive disease; 5 developed a progression, and 18 metastases were stable or showed a partial remission (Supplementary Table 3).
Circulating lymphocyte kinetics.An increase in the absolute counts of circulating CD8 + CTLs at FU1 compared to baseline (the primary study endpoint) was detected only in 12% of the patients in the SBRT and 20% of the patients in the cSBRT group.
Association between circulating lymphocyte changes and PFS.In landmark analyses, the median PFS from the rst follow-up in the SBRT group was 19 months for the patients without an increase in the CD8 + CTLs and was not reached in the patients with an increase in the CD8 + CTLs (HR = 0.484, 95% CI 0.060-3.927,p = 0.49).In the cSBRT group, the median PFS from the rst follow-up was 6 vs 3 months in patients without vs with an increase in the CD8 + CTLs (HR 0.944, 95% CI 0.233-3.818p = 0.93).
Changes in lymphocyte subsets.There was a decrease in the mean absolute counts of CD8 + CTLs and CD4 + T-cells compared to pre-treatment values, which was more prominent at the end of treatment in the SBRT group compared to the cSBRT group (Figs.1a-b, 3a-b, Table 2, Supplementary Fig. 3a-b Supplementary Table 4).This nding is consistent with the changes seen after SBRT in patients with early-stage primary lung tumors 10 .Furthermore, the standardized change from baseline was greater in the group with concurrent systemic treatment (cSBRT group) compared to the SBRT group until the rst follow-up (Supplementary Table 4).
In contrast, the fractions of proliferating CD8 + and CD4 + T cells T-cells were signi cantly increased only at the end of treatment in patients who received SBRT with systemic therapy (Fig. 3c-e).The increase in the fraction of proliferating T-cells was delayed and transient (at FU1 only) in the cSBRT group patients who received ICB and increased at the end of treatment only in CD8 + T-cells in patients who received additional chemotherapy (Fig. 3k-l, Table 2).
Most importantly, the effect size of standardized changes from baseline in proliferating T-cells was large in the SBRT group but only moderate to large in the cSBRT group and considerably lower compared to the SBRT group (Supplementary Table 4).In the cSBRT group, there was no difference in the standardized changes from baseline in proliferating T-cells between patients receiving ICB and patients receiving other systemic treatments at the end of treatment compared to baseline (CD8 + T-cells 0.89 vs 0.90, CD4 + Tcells 0.49 vs 0.45, respectively).
Changes in circulating lymphocyte phenotypes after SBRT.Median uorescence intensity (MFI) of PD-1 immunostaining was also higher at the end of treatment and at FU1 for the CD8 + T-cells, indicative of an increased expression level which was signi cant in the SBRT group (Figs.1h, 3h, Supplementary Fig. 3h, Table 2) but the effect size based on the standardized change from baseline was small (Supplementary Table 4).Additionally, the fractions of T-cells expressing the activation marker IFN-γ increased from baseline through to FU2 and the change was signi cant at the end of treatment in the cSBRT group, with a small effect size based on the standardized change from baseline (Supplementary Table 4).The fractions of T-cells expressing the activation marker IL-17A increased at the end of treatment (CD4 + Tcells) and FU2 (CD4 + T-cells) in the SBRT group with a small effect size (Figs.2a-c, 4a-c, Supplementary metastatic disease.We found promising PFS rates in all groups, with the longest in the cSBRT /ICB group. The primary endpoint of our study was not met, as only 12% of the patients in the SBRT and 20% of the patients in the cSBRT group experienced an increase in the absolute counts of CD8 + T-cells at FU1 compared to baseline.Furthermore, our ndings con rm the transient lymphopenia effects of SBRT in lung metastatic disease, as previously reported for SBRT in early-stage NSCLC patients 10 .This effect was more pronounced in patients who received concurrent systemic treatment.Importantly, we also validated our previous nding that SBRT-induced lymphopenia is followed by an increased T-cell proliferation, which may include tumor-speci c T-cells [14][15][16][17] , both with or without systemic treatment.Although the effects were generally similar in both groups, the effect size of standardized changes from baseline was considerably higher in the group without concurrent systemic treatment.Interestingly, the increased T-cell proliferation was maintained at the rst and second follow-ups in the patients without systemic treatment (SBRT group), suggesting that the timing of systemic treatment is critical for maximizing the immunerelated effects of SBRT.Importantly, concurrent systemic treatment did not lead to an augmentation of this systemic immunomodulatory effect.Previous studies in mice have shown that the therapeutic effect of SBRT combined with anti-PD1 was considerably reduced when immunotherapy was initiated before irradiation 18 .
Additionally, in patients with oligo progressive disease, without concurrent systemic treatment (SBRT group), T-cells seemed to be more exhausted at baseline than in those with systemic treatment.In patients with oligometastatic disease, SBRT has the potential to both reduce tumor burden and promote T-cell responses against micrometastases 19,20 .We did not observe any abscopal effects, in patients with untreated but stable metastases.Furthermore, the median PFS was longer in the SBRT group than that in the cSBRT group, but this result is confounded by the different tumor histologies, as most solid tumors are highly heterogeneous and evolve dynamically during treatment 21 .Interestingly, PFS was signi cantly longer in patients with concurrent ICB treatment.However, we did not nd any additional impact of ICB or chemotherapy on the systemic immune responses at the time-points evaluated in this study.
As previously described, the increased proliferation of CD8 + and CD4 + circulating T-cells occurred only in the patients treated with 10 Gy or less per fraction 10 .This effect could be attributed to the upregulation of three-prime repair exonuclease 1 (TREX1) 6,22 .These results were validated for SBRT in the metastatic disease setting.
Our study has several limitations.Due to the different duration of SBRT regimens, post-treatment evaluations were not time matched.Furthermore, we included patients with different histologies, which carry a differential prognosis.Given the heterogeneity in tumor histologies, these SBRT outcomes need to be validated in a disease-speci c manner.Nevertheless, the parameters of radiation dose and fractionation scheme, administration schedule, and target volume are all likely to have a crucial in uence on the ability of radiotherapy to elicit immunostimulatory effects that can be exploited with immunotherapies (notably ICIs) towards superior clinical e cacy 22 .This study addressed some of these important open questions.
In conclusion, our study shows that ablative SBRT leads to transient lymphodepletion but a signi cant increase in the fraction of proliferating CD4 + and CD8 + circulating T-cells after treatment in lung metastatic disease, consistent with our ndings in primary lung cancer.The increase was present at the end of treatment and follow-up but was more pronounced in the SBRT group than in the cSBRT group, and only when using doses of 10Gy or less per fraction.These data might help future studies of optimal integration of ICBs with SBRT in oligo-metastatic or oligo-progressive disease, for example starting after the end of SBRT with doses of less than 10Gy per fraction.Tables Tables 1 and 2 are available in the Supplementary Files

Figures
Figures

Figure 4 Expression
Figure 4