import heapq
from collections import deque

INF = 10**18
class Dijkstra():
    # 全ての辺の距離が1のとき
    def bfs(self, graph, start, goal=-2):
        N = len(graph)
        is_visited = [False]*N
        is_visited[start] = True
        dists = [INF]*N
        dists[start] = 0
        
        que = deque([(start, 0)])
        while que:
            p, d = que.popleft()
            for next in graph[p]:
                if not is_visited[next]:
                    que.append((next, d+1))
                    dists[next] = d+1
                    is_visited[next] = True
                    if goal == next:
                        return dists
        return dists

    def dijkstra(self, graph, start):
        N = len(graph)
        dists = [INF for _ in range(N)]
        dists[start] = 0
        is_done = [False for _ in range(N)]
        hp = []
        # 経路復元が必要な場合
        prev = [-1 for _ in range(N)]
        heapq.heappush(hp, (dists[start], start))

        while hp:
            min_dist, min_v = heapq.heappop(hp)
            if is_done[min_v]:
                continue
            is_done[min_v] = True
            # 隣接リスト形式でコストが無いなら、ここのcostを削除することに注意
            for to, d in graph[min_v]:
                # 隣接リスト形式でコストが無いなら、ここのコストを1に変更することに注意
                dist = dists[min_v] + d
                if  dist < dists[to]:
                    dists[to] = dist
                    prev[to] = min_v
                    heapq.heappush(hp, (dists[to], to))
        return dists, prev

    # 後ろから順になっている事に注意
    def get_path(self, prev, end):
        path = []
        p = end
        while prev[p] != -1:
            path.append(p)
            p = prev[p]
        path.append(p)
        return path
    

N, M = map(int, input().split())
graph = [[] for _ in range(N)]

for _ in range(M):
    ui, vi = map(int, input().split())
    ui -= 1
    vi -= 1

    graph[ui].append((vi, 1))

dijk = Dijkstra()
dist_from_0, _ = dijk.dijkstra(graph, 0)
dist_from_N1, _ = dijk.dijkstra(graph, N-2)
dist_from_N, _ = dijk.dijkstra(graph, N-1)

ans = INF
ans = min(ans, dist_from_0[N-2] + dist_from_N1[N-1 ]+ dist_from_N[0])
ans = min(ans, dist_from_0[N-1] + dist_from_N[N-2] + dist_from_N1[0])
if ans == INF:
    print(-1)
else:
    print(ans)