It depends on the requirement of the method. For example, if you want to detect related substances that are present in 0,05% you may need to raise the sample concentration to where you can reliably, accurately and repeatably determine those related substances.
That is why mostly the sample concentration is higher for related substances than for assay of main substance.
It is likely (sometimes) that the RS will have the same structure and solubility as the API since it is derived from the API. So I would use the same extraction method 1st. However, be sure you run a spike as well to verify your recovery.
Agreed with Nina Žigart . Moreover if your molecule and desired impurities are having good chromophoric groups, then it will give you enough response on UV detector HPLC. However 1000 ppm concentration will be optimum to get the desired impurity profile for your RS method. For non-chromophoric your molecule and impurities will give low response on UV, then you need to raise the sample concentration or if required any alternate detector.