Tumor　heterogeneity　is　one　major　reason　for　unpredictable　therapeutic　outcomes,　while　stratifying　therapeutic　responses　at　an　early　time　may　greatly　benefit　the　better　control　of　cancer.　Here,　we　developed　a　hybrid　nanovesicle　to　stratify　radiotherapy　response　by　activatable　inflammation　magnetic　resonance　imaging　(aiMRI)　approach.　The　high　Pearson＇s　correlation　coefficient　R　values　are　obtained　from　the　correlations　between　the　T-1　relaxation　time　changes　at　24-48h　and　the　ensuing　adaptive　immunity　(R=0.9831)　at　day　5　and　the　tumor　inhibition　ratios　(R=0.9308)　at　day　18　after　different　treatments,　respectively.　These　results　underscore　the　role　of　acute　inflammatory　oxidative　response　in　bridging　the　innate　and　adaptive　immunity　in　tumor　radiotherapy.　Furthermore,　the　aiMRI　approach　provides　a　non-invasive　imaging　strategy　for　early　prediction　of　the　therapeutic　outcomes　in　cancer　radiotherapy,　which　may　contribute　to　the　future　of　precision　medicine　in　terms　of　prognostic　stratification　and　therapeutic　planning.　Non-invasive　approaches　to　stratify　early　responses　to　therapy　are　of　great　value　to　improve　cancer　patient　management.　Here　the　authors　design　a　hybrid　nanovesicle-based　activatable　inflammation　magnetic　resonance　method　for　the　early　prediction　of　response　to　radiotherapy.