Circulating Cytokines and Angiogenic Markers as Biomarkers for Pazopanib in Metastatic Renal Cell Carcinoma

  • Date: 14 Aug 2012
  • Author: Joaquin Mateo
  • Affiliation: Drug Development Unit. The Royal Marsden NHS Foundation Trust & Institute of Cancer Research Sutton, United Kingdom
  • Link: Read the original article
  • Topic: Genitourinary cancers

Increasing understanding of renal cancer biology has led to identification of the vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF) and mammalian target of rapamycin (mTOR) as rational targets for development of novel therapies.  Since the first approval of Sorafenib by the FDA in 2005, several drugs as Bevacizumab, Sunitinib, Temsirolimus or Axitinib have broad the range of options. Pazopanib, an oral multikinase inhibitor (targeting VEGF receptors 1,2,3, PDGF receptors α and β and c-KIT) has been recently approved after a randomised, double-blind, placebo-controlled, phase III trial, showing an improvement in progression free survival of 9.2 months versus 4.2 months (P < 0.0001)(1). Identification of biomarkers to predict response is necessary to optimise personalised selection of therapies.

In this paper, Tran et al assessed the prognostic and/or predictive value of several cytokines and angiogenic factors (CAF) in mRCC treated with Pazopanib, using peripheral blood samples from patients who participated in the phase II (single arm) and III (randomised, placebo-controlled) clinical trials. All patients were sampled twice before receiving treatment and the mean values of plasma protein concentration of the several CAF were used, in order to reduce the impact of intra-patient and assay variability. CAF levels were dichotomised (high/low) using the median level in the population as cutout point and correlated with tumour shrinkage, progression-free survival (PFS) and overall survival.
A three-stages approach was used. First (“screening”), samples from the third of best and worst responders among the 225 participants in the single-arm phase II trial were tested for up to 17 CAF selected in the basis of previous published works. Samples were tested in two different biochips to allow large screenings with low amounts of plasma. Five candidates CAF showed correlation with PFS and/or tumour shrinkage: Hepatocyte Growth Factor (HGF), Interleukin-6 (IL6), Interleukin-8 (IL8), Tissue Inhibitor of Metalloproteases-1 (TIMP-1) and E-selectin. These were selected for further study together with VEGF, carbonic anhydrase IX, collagen IV and Osteopontin, added based on the results of other studies, specially a recent publication by the same group from the Sorafenib+Interferon trial (2).

In a second step (“confirmation”), all the available samples of the phase II trial participants were tested for these 9 CAF by ELISA and/or SearchLight Protein array. Lack of sufficient plasma volume hindered homogeneous use of both assays for all samples. Low concentrations of HGF, IL6, IL8, VEGF, TIMP-1 and Osteopontin were associated with responses. With the exception of VEGF levels, they all also correlated with PFS. High levels of E-selectin showed no correlation with tumour shrinkage but with extended PFS.

As this “confirmation” phase included the subset of samples already assessed in the “screening” phase, an external validation was required, and so a third “validation” phase considered samples from the randomised, placebo-controlled phase III trial of Pazopanib in mRCC (n=344 samples, 79% of 435 trial patients).  The analysis of the placebo-arm samples allowed assessing also predictive value of these CAF. The SearchLight protein array was the assay used in this last step. IL8 and osteopontin were prognostic on both the treatment and placebo groups, while IL6, HGF and TIMP-1 showed mismatched results between both groups. Only IL6 levels demonstrated a significant interaction on the response to treatment, being considered as the only CAF with predictive value for Pazopanib treatment.

An exploratory analysis compared CAF values and commonly used prognostic scores. The authors concluded that IL6, IL8 and Osteopontin were stronger prognostic markers than actual clinical classifications. However, it would have been interesting to integrate CAF levels and clinical criteria on a multivariate model.
Finally, they assessed the value of a protein expression signature based on these CAF, but it gives no more value than the individual factors. This group previously described this signature, based on unsupervised clustering of several markers, for predicting benefit of adding IFN to Sorafenib, identifying two subgroups with predominance of “angiogenic” or “inflammatory” factors. However, the value of the signature in the Pazopanib setting is unclear.In a second step (“confirmation”), all the available samples of the phase II trial participants were tested for these 9 CAF by ELISA and/or SearchLight Protein array. Lack of sufficient plasma volume hindered homogeneous use of both assays for all samples. Low concentrations of HGF, IL6, IL8, VEGF, TIMP-1 and Osteopontin were associated with responses. With the exception of VEGF levels, they all also correlated with PFS. High levels of E-selectin showed no correlation with tumour shrinkage but with extended PFS.

As this “confirmation” phase included the subset of samples already assessed in the “screening” phase, an external validation was required, and so a third “validation” phase considered samples from the randomised, placebo-controlled phase III trial of Pazopanib in mRCC (n=344 samples, 79% of 435 trial patients).  The analysis of the placebo-arm samples allowed assessing also predictive value of these CAF. The SearchLight protein array was the assay used in this last step. IL8 and osteopontin were prognostic on both the treatment and placebo groups, while IL6, HGF and TIMP-1 showed mismatched results between both groups. Only IL6 levels demonstrated a significant interaction on the response to treatment, being considered as the only CAF with predictive value for Pazopanib treatment.

Finally, they assessed the value of a protein expression signature based on these CAF, but it gives no more value than the individual factors. This group previously described this signature, based on unsupervised clustering of several markers, for predicting benefit of adding IFN to Sorafenib, identifying two subgroups with predominance of “angiogenic” or “inflammatory” factors. However, the value of the signature in the Pazopanib setting is unclear. predictive value for Pazopanib treatment.

An exploratory analysis compared CAF values and commonly used prognostic scores. The authors concluded that IL6, IL8 and Osteopontin were stronger prognostic markers than actual clinical classifications. However, it would have been interesting to integrate CAF levels and clinical criteria on a multivariate model.
Finally, they assessed the value of a protein expression signature based on these CAF, but it gives no more value than the individual factors. This group previously described this signature, based on unsupervised clustering of several markers, for predicting benefit of adding IFN to Sorafenib, identifying two subgroups with predominance of “angiogenic” or “inflammatory” factors. However, the value of the signature in the Pazopanib setting is unclear.

Overall, this paper represents another step in the validation of CAF as a tool in clinical-decision making in mRCC, but this is quite far yet. With several drugs of similar mechanism of action, there is not yet an agreement about whether we have to look for common predictive factors to a family of drugs (and then analyse them in comparative face-to-face trials) or selective biomarkers should be validated independently for each drug. The current scenario of retrospective analyses of biomarkers in slightly different settings and different drugs makes difficult to compare results of different studies. Lack of confirmed prognostic value of VEGF levels is contradictory in this study compared to others previously reported (3, 4) This reinforces the idea that assays to evaluate CAF are yet to be improved to be reliable and reproducible, but this study is an advance in this direction. High intra-patient variability of CAF levels is a handicap for their assay validation.

In this particular study, only baseline CAF values are analysed. There is evidence of modulation of VEGF-related proteins as well as IL6 and IL8 in mRCC (and neuroendocrine tumours) after exposure to anti-VEGF tyrosine kinases inhibitors. Therefore, final results of the analysis of changes in these CAF after treatment and the predictive value of this changes is awaited. Regarding the methodology, it worth to mention the use of several assays during the study. Despite this may lead to some confusion and risk the reproducibility of the work, the authors present in the supplemental data good correlation among the results of the several platforms, except for IL8 and TIMP-1.

In summary, the paper reflects the idea that CAF may potentially become important for prognostic assessment in advanced mRCC and personalised treatment selection, either on their own of in conjunction with current clinical criteria, but further studies and methodology standardisation are required.

Reference

Tran, H. T., Liu, Y., Zurita, A. J., Lin, Y., Baker-Neblett, K. L., Martin, A.-M., Figlin, R. a, et al. (2012). Prognostic or predictive plasma cytokines and angiogenic factors for patients treated with pazopanib for metastatic renal-cell cancer: a retrospective analysis of phase 2 and phase 3 trials. The lancet oncology, 13(8), 827–837.

References for further reading

1.    Sternberg CN, Davis ID, Mardiak J, et al. Pazopanib in locally advanced or metastatic renal cell carcinoma: results of a randomized phase III trial. J Clin Oncol 2010; 28: 1061–68.
2.    Zurita AJ, Jonasch E, Wang X, et al. A cytokine and angiogenic factor (CAF) analysis in plasma for selection of sorafenib therapy in patients with metastatic renal cell carcinoma. Ann Oncol 2012; 23: 46–52.
3.    Deprimo, S. E., Bello, C. L., Smeraglia, et al. Circulating protein biomarkers of pharmacodynamic activity of sunitinib in patients with metastatic renal cell carcinoma: modulation of VEGF and VEGF-related proteins. Journal of translational medicine, 2007, 5, 32.
4.    Rini, B. I., Michaelson, M. D., Rosenberg, J. E., Bukowski, R. M., Sosman, J. a, Stadler, W. M., Hutson, T. E., et al. Antitumour activity and biomarker analysis of sunitinib in patients with bevacizumab-refractory metastatic renal cell carcinoma. Journal of clinical oncology 2008, 26(22), 3743–8. doi:10.1200/JCO.2007.15.5416
5.    Zurita, A. J., Jonasch, E., Wu, H et al. Circulating biomarkers for vascular endothelial growth factor inhibitors in renal cell carcinoma. Cancer, 2009, 115(10 Suppl), 2346–54.

You should Sign in to post a comment