The gravitational lensing method for determining the Hubble constant was first introduced by Refsdal (Refsdal 1964, MNRAS, 128, 307) and provides a means of determining a global value for the Hubble constant based on the measurement of the light travel delay between different images of a gravitationally lensed (GL) system (time-delay). Together with a model that describes the gravitational potential of the lens, the time-delay defines the geometrical dimensions of the lens system and may, therefore, lead to the determination of the Hubble constant. The main difficulty in measuring time-delays is that the brightness of each image has to be carefully monitored over several periods of the time-delay and the quasar has to show sufficient variability over time scales smaller than the time-delay. X-ray observations of several GL quasars, show strong variability over time scales of hours to days. We recently reported the first detection of time-delayed flares in single X-ray observations of the gravitationally lensed quasar PG 1115+080 performed with the Chandra and XMM-Newton observatories. We found a short time-delay between images A1 and A2 of PG 1115+080 of dt_A1A2 = 0.149 +/- 0.006 days. This in combination with recent constraints on the mass fraction and slope of the dark matter component of the lensing galaxy (Treu & Koopmans 2002, MNRAS, 337, L6) then restricts the allowable set of lens models for PG 1115+080 and yields a tighter limit on the Hubble constant via the lensing method of H_0 = 67 (-8,+13) +/- 3 km/s/Mpc (random + systematic errors). Alternatively, one may use time-delays with an independently measured value of H_0 (e.g., WMAP, HSTKP) to constrain the dark mater distribution of the lensing galaxy.