External beam radiotherapy (RT) has constantly advanced through accelerator research and development. Translational research aims to streamline the transition to specialized radiation sources for improved patient care. The "FLASH" effect, triggered by fast dose delivery (<100 ms) at high dose rates (≈40 Gy s⁻¹), shows promise by offering unaltered tumor response with reduced normal tissue toxicity. This allows dose escalation and minimizes organ motion during treatment. However, the transition to FLASH RT requires detailed *in vivo* studies on normal and tumor tissue response, considering dose per bunch, bunch repetition frequency, bunch dose rate, mean dose rate, fractionation, and total irradiation time. Such investigations require research accelerator infrastructure for small animal studies, ideally capable of delivering various dose and dose rate schemes. While electron and X-ray FLASH studies are underway, research on proton RT's dose-rate effects is inconclusive due to limited accelerators capable of delivering ultrahigh dose rates. This study investigates laser-plasma acceleration (LPA) as a potential solution.

Several studies have explored the benefits of ultrahigh dose rate radiation therapy, particularly the FLASH effect. Favaudon et al. (2014) demonstrated increased differential response between normal and tumor tissue in mice using FLASH irradiation. Subsequent studies by Bourhis et al. (2019) and Vozenin et al. (2019) discussed the clinical translation of FLASH radiotherapy and its biological benefits. However, questions remain about the optimal dose rate and the potential for similar effects with proton therapy. Existing studies using LPA protons have been limited to *in vitro* experiments due to limited proton energies. Recent advancements in petawatt-class lasers now allow for the generation of proton bunches with sufficient energy for *in vivo* studies, but significant technological challenges remain in creating a stable and reliable platform for such research.

This study used a mouse model with spherical tumors (~3 mm diameter) grown superficially on a mouse ear. A cylindrical planning target volume (PTV) of 5 mm diameter and 4 mm depth (water equivalent) was defined. A homogeneous dose of 4 Gy was prescribed, with a minimum mean dose rate of 1 Gy min⁻¹. Irradiation was performed at the Draco PW laser, which delivers up to 18 J in 30 fs pulses on the target. Protons are emitted from plastic foils, and the spectrum is tailored using a pulsed two-solenoid beamline, apertures, and scatterers to achieve the desired depth-dose homogeneity. A time-of-flight (TOF) spectrometer monitors the proton spectrum for the PTV, providing calibrated on-shot depth-dose profile prediction. An energy selection aperture (ESA) helps to shape the spectrum for single-shot depth-dose homogenization. Lateral beam confinement and homogeneity were achieved using a final beam aperture and scatterers. A multi-shot irradiation scheme was employed, with single-shot doses adjusted by controlling laser input energy. A transmission ionization chamber (IC) and radiochromic film (RCF) were used for dosimetry. Reference irradiations were conducted at the University Proton Therapy Dresden (UPTD) using a clinical proton source. Human head and neck squamous cell carcinoma tumor cells were injected into mouse ears. Animals were allocated for treatment when a tumor of ~3 mm developed. Tumor growth was followed for up to 120 days after irradiation. Control groups included sham-irradiated and non-irradiated mice.

The laser-driven proton source demonstrated long-term stability over a two-year period, with daily available cut-off energy consistently near 60 MeV. Daily average dose delivery parameters (dose per shot, depth-dose inhomogeneity, lateral dose inhomogeneity) remained within the targeted ranges. Shot-to-shot stability was also high. The prescribed dose of 4 Gy was precisely delivered at the laser-driven source using a multi-shot scheme, achieving the required dose homogeneity. The applied proton tumour dose values at both facilities (Draco PW and UPTD) were comparable (~3.9 Gy). Both dosimetry methods (IC and dRCF) agreed well, and dose inhomogeneity was comparable between the two facilities. Tumor growth curves showed a clear radiation-induced growth delay in mice irradiated with both the laser-driven and clinical proton beams, though sample size was limited. The platform is capable of delivering single-shot doses exceeding 20 Gy, homogeneously distributed over millimeter-scale volumes at ultrahigh dose rates (10⁶ Gy s⁻¹).

This study successfully established a proton LPA research platform for small animal studies. The platform met all requirements for accelerator readiness, three-dimensional tumor-conform dose delivery, dosimetry, and radiobiological protocols. The results demonstrate the feasibility of using laser-driven proton sources for *in vivo* radiobiological research. The comparable results obtained at both the laser-driven and clinical proton facilities validate the platform's capabilities. The high single-shot dose delivery capability opens exciting possibilities for investigating the FLASH effect with protons and exploring other radiobiological questions.

This *in vivo* pilot study demonstrates the successful establishment of a laser-driven proton research platform for small animal studies. This platform enables the delivery of precisely controlled, three-dimensional doses at ultrahigh dose rates, making it suitable for investigating the FLASH effect and other radiobiological phenomena. Future studies with larger sample sizes will be crucial for further exploring the therapeutic potential of laser-driven protons in cancer therapy.

The pilot study's primary limitation is the relatively small sample size (61 mice), which limits the statistical power of the radiobiological analysis. Further studies with increased sample sizes are needed to confirm the observed radiation-induced tumor growth delay and to definitively determine the efficacy of LPA proton irradiation compared to conventional proton beams. The study focused on a single dose point (4 Gy), and future research should investigate the dose-response relationship.