The motivation behind providing Location Based Services (LBS) to mobile clients is due to increase in number of mobile clients. In LBS data is provided to the mobile user's application based on users location. It has got many challenging problems for location-dependent query processing and management of data. Data caching at mobile clients while user is moving is an efficient mechanism to enhance data availability and reduces data access cost. Due to the limitation of client cache, the cache replacement technique can be used to find a proper subset of items for eviction is important. Several policies has been proposed in literature. Prioritized Predicted Region based Cache Replacement Policy (PPRRP) considers spatial as well as temporal aspects for replacement. PPRRP does not focus on data size, which can be important factor for performance improvement. We propose an improved PPRRP cache replacement policy for mobile environment. It uses a predicted region based cost function to select an item for eviction from cache. iPPRRP data replacement is based on valid scope, access frequency, data distance and data size of data item. This makes the policy adaptive to client's movement pattern.