Author's response to reviews Title: Interstitial fluid pressure, vascularity and metastasis in ectopic, orthotopic and spontaneous tumours. Authors:

Background: High tumour interstitial fluid pressure (IFP) has been adversely linked to poor drug uptake in patients, and to treatment response following radiotherapy in cervix cancer patients. In this study we measuredcompared IFP values in an array of murine and xenograft models, spontaneously arising or transplanted either intramuscularly (i/m) or orthotopically and analysed their relationship to tumour vascularity and metastatic spread.. Methods: KHT-C murine fibrosarcoma, ME180CaSki, Me180 and SiHa human cervix carcinoma were grown either intramuscularly (i/m),i/m, sub-cutaneouslysub-cutaenously (s/c) or orthotopically. The Polyoma middle-T (MMTV-PyMT) transgenic spontaneous mammary tumours were studied either as spontaneous tumours or following orthotopically or i/m transplantation . model was included for comparison. IFP was measured in all tumours using the wick-in-needle method. Spontaneous metastasis formation in the lungs or lymph nodes was assessed in all models. An immunohistochemical analysis of tumour hypoxia, vascular density, lymphatic vascular density and proliferation was carried out in ME180Me180 tumours grown both i/m and orthotopically. Blood flow was also assessed in the ME180Me180 model using high-frequency micro-ultrasound functional imaging. Results: Tumour IFP was heterogeneous in all the models irrespective of growth site: KHT-C i/m: 2-42 mmHg, s/c: 1-14 mmHg, ME180:CaSki: i/m 6-63 mmHg, s/c: 1-20 mmHg, cervix 1-18 mmHg, Me180: i/m 5-68 mmHg, cervix 4-21 mmHg, SiHa: i/m 20-56 mmHg, cervix 2-26 mmHg, MMTV-PyMT: i/m: 13-45 mmHg, spontaneous 2-20 mmHg and transplanted 2-22 mmHg. Additionally, there was significant variation between individual tumours growing in the same mouse, and there was no correlation between donor and recipient tumour IFP values. Metastatic dissemination to the lungs or lymph nodes demonstrated no correlation with tumour IFP. Tumour Furthermore, immunohistochemical analysis of tumour hypoxia, proliferation, and lymphatic or blood vessel density also showed nodid not demonstrate any relationship with tumour IFP. SpeckleHowever, speckle variance analysis of ultrasound images MS # 1765358339136524 Version 2; Response to reviewers comments Page 9/40 showed no differences in vascular perfusion between ME180Me180 tumours grown i/m versus orthotopically despite differences, suggesting that the functionality of the vasculature is a key factor in IFP.the inter-model heterogeneity. Conclusions: Our studies across a range of different tumour models showed substantial heterogeneity in tumour IFP, suggesting differences in the vascular development and interstitial fluid dynamics in the individual tumours. The results demonstrate a strongexistence of a stochastic aspect to tumour IFP development, notably irrespective of previous molecular interactions was further enforced by the variation apparent betweenin different tumours within the same animal and the lack of correlation between donor and recipient tumours. Given the marked variability in IFP as a function of tumour type and implantation site, future studies may be more effective using tumours grown orthotopically, which are more relevant to both hosttumour specific interactions and clinical interpretation. MS # 1765358339136524 Version 2; Response to reviewers comments Page 10/39