Gravitational lensing by massive galaxy clusters is a powerful tool for the discovery and study of high redshift galaxies, including those at z>=6 likely responsible for cosmic re-ionization. Pello et al. recently used this technique to discover a candidate gravitationally magnified galaxy at z=10 behind the massive cluster lens Abell 1835 (z=0.25). We present new Keck (LRIS) and Spitzer Space Telescope (IRAC) observations of the z=10 candidate (hereafter #1916) together with a re-analysis of archival optical and near-infrared imaging from the Hubble Space Telescope and VLT respectively. Our analysis therefore extends from the atmospheric cut-off at lambda_obs=0.35um out to lambda_obs=5um with Spitzer/IRAC. The z=10 galaxy is not detected in any of these data, including an independent reduction of Pello et al.s discovery H- and K-band imaging. We conclude that there is no statistically reliable evidence for the existence of #1916. We also assess the implications of our results for ground-based near-infrared searches for gravitationally magnified galaxies at z>~7. The broad conclusion is that such experiments remain feasible, assuming that space-based optical and mid-infrared imaging are available to break the degeneracy with low redshift interlopers (e.g. z~2-3) when fitting spectral templates to the photometric data.