This is just my opinion, but I hope that it will hold some insight for you.
In regards to the current meta and possible counters to it, there are no 100% counters.
One of the biggest variables in every game of League you play is the enemy teams skill/teamwork. No two teams play the same and no two players will respond to different situations in the same manner.
Many players have become accustomed to the current meta game of sending-
1 Bruiser type solo Top
1 AP caster/nuker/assassin Mid
1 Jungle
and
1 Ranged AD carry and 1 Support Bot
This team composition has been found to be not only the most efficient, but to provide in a typical game the highest chance for success.
This does not however mean it will always result in a win for the team that uses this format.
One of the reasons for this is that with set lane compositions, you often know who you will be facing in lane before you ever get there. This allows for strategies and tactics for each lane to be devised before you ever encounter your enemy in the lane.
Teams that decide to try something different, whether its sending that AP caster/nuker bot with your AD carry, not having a jungle so that your bruiser top is no longer going to get zoned out by that counter pick, sending your tank mid, all of these things can not only confuse the enemy team, but they can put them out of their comfort zone.
Being out of that comfort zone, not knowing how to deal with that dual top lane or that tank that you just can't seem to do anything to mid, can majorly screw with a player and a team's mindset.
Counters to the meta that I have seen be effective were not always necessarily fantastic teams that just switched up where a specific role goes. Part of the game of League of Legends is the mental aspect to it. A team that can confuse, agitate, or otherwise just throw off the shackles of what the enemy expects them to do, can win games regardless of filling those typical meta roles.
I have personally seen teams of tanks, teams of assassins, teams of AD carries and teams of champions that would typically be viewed as supports just roll the enemy team. Not because their team composition was better, but because they were able to change the flow of the game. They were able to force the enemy team to deal with situations that they normally would not find themselves in, fighting champions that they may very well not know how to play against.
The biggest thing with these teams, regardless of their team composition is that they are able to take control of the game. When they gained an advantage, they kept it. They push, gank, go for dragon/baron when they can take it, ward and communicate with their team. They "know" what they need to do to win, and they also know that they often times have to win before the game goes on to long.
In conclusion, although there is no "hard" counter to the current meta, with proper knowledge of the game, a little bit of innovation and a whole lot of teamwork, it is "possible" to win with something other than the current meta. (You just have to try!)
The 'Blue Ezreal' build focuses around poke items, compared to the typical ADC build which would focus more on autoattack damage and less around poking with Q. Both builds are high damage, but differ on the situation.
If your team is designed around fast engages, the Blue Ezreal isn't as good as the typical ADC Ezreal: As stated before, Blue Ezreal shines when poking and/or chasing (due to the slow of gauntlet)
Many people get Elder Lizard prior to Tears due to it's decent damage and on-hit feature. Although getting Tears first to start stacking it early may seem like a good idea, it sometimes can cause you to become too behind in lane (depending on matchup) since it provides no damage or health/sustain, and therefore a lot of players opt to rush the Elder Lizard first, since it provides an easier laning phase, supplying you with HP regen and damage, for better damage trading with the opponent. It also builds from 2 longswords, so you can grab them earlier to bully your opponenets.
That being said, it is possible to get Tears first to start stacking it early, but, you might find yourself struggling in lane more, depending on matchup.
Best Answer
You cannot. Well, you can but it's risky, risky enough that it comes down to who is your ADC.
As you know by now (hence the question), two or more champions near a minion/champion death results in less experience gain. That is why a common tactic (not necessarly only in LCS) is to leave the ADC alone, in order for him to obtain more experience.
Well, for one there's Zilean's passive like @Dragonrage mentioned in the comments, but I suspect you want a more definite answer than "play x champion".
What you need to ensure is that your ADC will still be able to farm without you, and that he won't die without you. In other words, you need to ensure that:
There's also another situation which is jungling, depending on who your botlane is, you can sometimes go into the jungle and take a camp or two to gain some experience if you're zoned in lane. In this case both you and the adc would take a camp and get a small experience boost instead of gaining no exp at all. However this depends heavily on your own jungler's routes and what he's trying to do at that momment.
Finally the most important thing is: You are being useful. There is no point in roaming to let your adc get level advantage if you're getting even further behind. Let's say you roam and your ADC is level 6. Their botlane is level 5 and you're level 4. They are ahead because they will have their ultimates on a 2v1 situation, or on a 2v2 situation before you have your own ultimate ability. As such you should only do those roams if you're being productive (Taking Dragon, Ganking Lanes, Peeling on jungle skirmishes)