I'm not sure you've understood the processing model.
If you're driving things from an ASP page, then typically the ASP page invokes an XSLT transformation of an XML document to produce an HTML document. The HTML document is then sent to the browser. The user might click on a button. This typically causes an HTTP request to be sent to the server. This can't interact with the original ASP page, which has long since been deactivated. Instead, it must be processed from scratch. The way you control this is that when you generate the HTML page, you put enough information in the links so that when a URL is sent back to the server, the server has enough information to know what to do next.
I hope this makes things clearer.
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference