Unfortunately, not possible.

Workaround: Create a Materialized view with calculated data.

Run in the MongoDB Shell:

var prev = null;
db.ATest_prevback.find({}).sort({field:1}).forEach(function(curr){
    db.materialized_view.insert({curr:curr.field, prev:(prev ? prev.field : 0)})
    prev = curr;
})

Run "fake" lead and lag queries:

db.materialized_view.aggregate([
    {$match:{$expr:{$eq:["$curr", "$prev"]}}}
])

Starting from MongoDB v5.0+, it can be done by using $shift in $setWindowFields.

db.collection.aggregate([
  {
    "$setWindowFields": {
      "partitionBy": null,
      "sortBy": {
        "entered_date": 1
      },
      "output": {
        lag: {
          $shift: {
            output: "$Dashbo",
            by: -1,
            default: "Not available"
          }
        },
        lead: {
          $shift: {
            output: "$Dashbo",
            by: 1,
            default: "Not available"
          }
        }
      }
    }
  }
])

Here is the Mongo playground for your reference.