If you were to decide to attempt a marathon given the training level you've described, there's an interesting academic article that may be helpful in estimating your likelihood of finishing without injury. The paper is by Yeung, Yeung, and Wong, "Marathon finishers and non-finishers characteristics: A preamble to success," Journal of Sports Medicine and Physical Fitness, 41(Jun 2001):170-176.
The researchers surveyed 113 registered runners at a Hong Kong marathon, of whom 55 finished and 58 were unable to finish. The predictive variables that had overwhelming explanatory power (p < 0.005) as to whether or not a runner would finish were:
1) weekly training mileage -- the training distance, per week, of the average finisher was 52km (32mi) whereas for the average non-finisher, it was 9km (5mi);
2) length of the longest run accomplished during training -- 28km (17mi) for the average finisher vs. 5.4km (3.4mi) for the average DNF (did not finish); and
3) the runner's personal opinion of how much training was needed to do a marathon -- that is, how difficult did the runner think a marathon would be? The average finisher assessed 72km (44mi) per week as optimal marathon training whereas the average non-finisher thought that only 29km (18mi) per week would be sufficient.
Given your question it seems that all three predictive variables (in their current state) are pointing towards a prediction of non-finishing.
Your question also asked about what the risks were if you did decide to attempt the marathon at this point. The same paper includes a breakdown of the non-finishers, that is, why did the non-finishers not finish? Of the non-finishers, 43% dropped out of the race from exhaustion,
35% stopped at the 10km mark due to missing the time cut,
19% were injured,
2% were advised at a medical tent to stop, and
2% reported "other."
