Dynamic Top N calculation for time series visual

Hello,

I have created a dynamic Top N calculation which follows the DAX logic in an article linked here. I have also pasted this DAX at the bottom this message.

Whilst this measure works as intended, when used in a time series line graph it evaluates the top N selection at each data point. Instead, I want it to evaluate across the entire data range in the visualisation - the same way the Top N filter works when applied to a graph.

Please see example below:

Would be extremely grateful for assistance on this one -- have sunk considerable time in without any result! 

DAX from linked article above:

Top Orders | Cities = 

VAR N = 'Top N'[Top N Value]

-- Get the top N cities by total orders
VAR Top_Cities = 
    TOPN( 
        N, 
        ALLSELECTED('Cities'[City]), 
        [Total Orders] 
    )

-- Calculate the total orders for the Top N cities
VAR TopN_Orders = 
    CALCULATE(
        [Total Orders],
        KEEPFILTERS(Top_Cities)
    )

-- Calculate total orders for all selected cities
VAR All_Orders = 
    CALCULATE(
        [Total Orders],
        ALLSELECTED(Cities)
    )

-- Calculate the total orders for cities that are not in the Top N
VAR Other_Orders = 
    All_Orders - CALCULATE([Total Orders], Top_Cities)


VAR Result = 
    SWITCH(
        TRUE(),
        NOT ISINSCOPE(Cities[City]), All_Orders,
        SELECTEDVALUE(Cities[City]) = "Others", Other_Orders,
        TopN_Orders
    )



RETURN 
    Result