The existence of really old people who've been smoking their entire lives is something you should expect to see if smoking is a strong statistical danger but not a guaranteed killer. It does not count as evidence against that theory.
By the way, I looked up lung cancer rates in the US and the UK, and they've both been going down quite significantly among men, who were at a higher risk to begin with, and rising slowly among women -- a trend which has tapered off more recently.
http://www.cdc.gov/cancer/lung/statistics/trends.htm
http://info.cancerresearchuk.org/cancerstats/types/lung/inci...
Looking at correlations between smoking rates and lung cancer rates in various countries, there's strong evidence that most of the mortality gap between men and women can be explained by the greater prevalence of smoking among men: