A cox regression analysis is an analysis you apply when you have time-to-event data which may be right censored. For example, if you look at time to death/mortality in a certain diagnostic group, but with incomplete follow-up (right censoring) due to subjects leaving the country; then a cox model would be ideal. Kaplan Meier graphs is a way of illustrating the accumulation of new events, as time passes.
There are several youtube tutorials on how to apply the model in SPSS. https://www.youtube.com/watch?v=bDwNkHGBixM
This is an excellent document to get a sense of survival analysis using SPSS: https://www.researchgate.net/profile/Maurice_Ekpenyong2/post/How_do_I_calculate_1-SD_for_a_co_variate_in_Survival_analysis/attachment/5c6eab43cfe4a781a581f057/AS%3A728740468367361%401550756675222/download/Survival+Analysis.pdf.