Do as you will. I flew the airplane for a thousand hours, and our company policy was nothing less than 4,000'.
You said your runway will be 4430 feet for a couple of years...does that mean it's that length now and they're lengthening it later?
Run your numbers. Bman has posted numbers here, and you can get them from the performance charts in your book. Raw distance numbers, yes, but realistic numbers...no. It's not a transport category airplane, and the performance numbers aren't gauranteed. My experience has been that the airplane doesn't do everything the Itallians like to say it does. it just doesn't. It's a great airplane, but I've watched average pilots use every inch of 4,000' runways on several occasions, and I've watched experienced pilots blow out the tires.
As I said, you do as you will. You can shoehorn the airplane into smaller spaces. But what about your takeoff distances when you lose one engine at V1...what about your abort distances?
Take accelerate-go distance over a 50' obstacle. 30 deg. C day, 2,000' field elevation, you're looking at 4,400' of published distance, assuming your departure profile is perfectly flown, and the numbers are correct. Don't count on it. Your field is closer to sea level, but at the same temperature at sea level, the book states 4,000' of distance. To abort the same takeoff, the book gives you 4,600' of distance with reverse. If you haven't tried an abort using significant reverse, you may not appreciate that control can quickly be lost...you most likely won't have reverse, or much reverse available. In other words, those distances will be longer than published. Further, if you're aborting with cold brakes, the distance will be longer yet, as the carbon brakes are poor when unheated.
A standard takeoff distance over a 50' obstacle under the same conditions, 30 deg C, at sea level, is 3,200'. Seems that's close to your runway length, isn't it? Again, bear in mind these aren't transport category numbers...no gaurantee that your airplane will meet them. Certainly don't count on your performance being better than the published numbers.
You can choose to ignore accelerate-stop distances, accelerate-go distances, and even climb gradients if you wish, but the data is right there, and it SHOULD be considered.
We always used 4,000' as a minimum number, and often 5,000 isn't enough. I'v flown the airplane in and out of fields nearly 10,000' in elevation, regularly out of places with that much density altitude; it does get off the ground, it does do it's job, but no matter where you fly, you need to be planning for te engine failure, not all-engine operation.