Fig 2: Impact of savolitinib on immune infiltration, tumor growth, and splenocyte migration in immunocompetent and immunodeficient SCLC models(A) Immune infiltration analysis from KP1 model after treatment. Frequency of tumoral M-MDSC from KP1 tumors across treatments. Control (n = 4), cisplatin + isotype IgG (n = 2), cisplatin + savolitinib (n = 4), cisplatin + anti-PD-L1 (n = 4), cisplatin + anti-PD-L1 → savolitinib + anti-PD-L1 (n = 4), and cisplatin + anti-PD-L1 + savolitinib (n = 4).(B) Immune infiltration analysis from RPP631 model after treatment. Frequency of splenic CD11b+ cells from the spleens of RPP631-bearing mice across treatments. Chemo-immunotherapy (CI) (n = 10) and chemo-immunotherapy plus savolitinib (CI + SV) (n = 10).(C) Tumor growth curve of the KP1 model in NSG mice treated with cisplatin, savolitinib, and anti-PD-L1. Cisplatin + isotype IgG (n = 3), cisplatin + anti-PD-L1 (n = 3), and cisplatin + anti-PD-L1 + savolitinib (n = 4).(D) Tumor growth curve of the KP1 model treated with cisplatin, anti-PD-L1, and either savolitinib or anti-Gr1. Cisplatin + anti-PD-L1 (n = 5), cisplatin + anti-PD-L1 + anti-Gr1 (n = 5), and cisplatin + anti-PD-L1 + savolitinib (n = 6).(E) Migration of murine splenocytes after 4 h of exposure to an HGF gradient. The data plot shows the percentage of total seeded cells that migrated. Data are represented as the mean ± SD (n = 3). ∗∗p < 0.01; ∗∗∗∗p ≤ 0.0001.(F) Representative images of CD11b immunofluorescence staining in migrated KP1 splenocytes after 4 h of exposure to an HGF gradient (100 ng/mL). The control condition was RPMI 1640 medium without HGF. DAPI was used as a counterstain. From left to right: representative images of the DAPI (in blue) and CD11b (in green) channels and the merged composition. Scale bar: 100 μm.In (A) and (B), data are shown as means ± SD, with dots representing individual samples. ∗p < 0.05. In (C) and (D), data are shown as means ± SEM. No statistically significant differences were observed between groups (ns). See also Figures S5–S7.

Fig 3: Evolution of serum cytokine levels at different stages of SCLC and the association of HGF with EMT features(A) Boxplots representing sHGF levels (pg/mL) across different time points: baseline (TP1), response (TP2), and progression (TP3). TP1 (n = 81), TP2 (n = 53), and TP3 (n = 40).(B) Boxplots representing sHGF levels (pg/mL) across different SCLC subtypes. SCLC-A (n = 19), SCLC-N (n = 6), SCLC-P (n = 3), and SCLC-non-A/N/P (SCLC-I) (n = 8).(C) Correlation analysis between the H-score of EMT markers (E-cadherin and vimentin) and baseline sHGF levels. Pearson’s correlation analysis was performed for both markers.In (A) and (B), boxplot shows the median as a thick line, the box highlighting the first and third quartile with the whiskers plotted with the Tukey method.See also Figure S8.

Fig 4: MMP-3 inhibits hepatocyte proliferation by inducing Met degradation. (A) Changes in the expression of proliferation markers in human hepatocytes treated with recombinant proteins and HGF. Recombinant DPT, LGALS1, TNC, MMP-2, or MMP-3 (100 ng/mL each) were administered to human primary hepatocytes from 2 different donors. (B) Evaluation of whether the HGF-induced increase in proliferation markers is altered by MMP-3 treatment in human hepatocytes. (C) Western blotting for c-Met downstream signaling molecules in human hepatocytes treated with HGF and MMP-3. Changes in protein expression were calculated as fold. (D) MMP-3 inhibitor restores expression of proliferation markers in MMP-3+HGF-stimulated human hepatocytes. (E) Expression of c-Met in human hepatocytes treated with HGF, MMP-3, and MMP-3 inhibitor. Data are presented as mean ± SD. Changes in protein expression were calculated as fold. The dot plot shows individual values. Comparisons between groups were analyzed using the Mann-Whitney U test. ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, ∗∗∗∗P < .0001. (F) Graphical summary: aPF-derived MMP3 facilitates shedding of c-Met, thereby limiting hepatocyte proliferation.

Fig 5: Hepatocyte proliferation is improved by Msln−/− PFs. (A) Livers were stained with anti-Ki67 Abs. Positive area was calculated as percent. The number of positive hepatocytes per high power field (HPF) was counted (10× and 20× objectives). (B) Expression of proliferation markers was analyzed in these mice using qRT-PCR, or (C) Western blotting for phospho-Akt, Akt, phospho-p38, and p38. (D‒G) A 70% partial hepatectomy (PH) or sham operation was performed using 12-week-old WT or Msln−/− mice (female, C57BL/6, n ≥ 4/group), and mice were sacrificed 3 days later. (D) Study design. (E) Livers were stained with anti-Ki67 Abs. Positive area was calculated as percent. The number of positive hepatocytes per high power field (HPF) was counted (20× objectives). (F) Expression of proliferation markers was analyzed in these mice using Western blotting for phospho-Akt, Akt, phospho-p38, and p38, or (G) qRT-PCR. (H) Primary hepatocytes were isolated from 8-week-old WT mice (male, C57BL/6) and seeded, and cultured in conditioned medium from WT aPFs or Msln−/− aPFs ± mouse recombinant HGF (for 48 hours). Expression of proliferation markers was assessed by qRT-PCR. Data are presented as mean ± SD. The dot plot shows individual values. Comparisons between 2 groups were analyzed using the Mann-Whitney U test. ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, ∗∗∗∗P < .0001.