It would be best to determine the cause of the damage to know if the fix is effective. If it is a cell transient during operation better filtering may help, if it is during connection, a fixture voltage limiting system may work and avoid cost in the product.
In the bottom up connection sequence mentioned above, connections are assumed to make and stay connected. When battery- is connected and remains, but the others connect and break sequentially bottom up, the VC10 voltage builds up. Assuming a 4V/cell voltage, 4V, then 6V, then 7.33V...
A zener/TVS on the lowest cell input will likely prevent the damage. You must pick a compromise between leakage and protection. You would like zero leakage at normal cell voltage and clamp below the device VC10 ABS MAX limit of 5V, but ideal diodes don't exist so you must select a compromise. A zener clamp at VC10 would be more certain, but since abs max is 5V and the pin must swing to an induced voltage above this in normal balancing operation, the designers/datasheet author have given us a negative margin for selecting a zener. If balancing is not used, a zener may be (more easily) placed on VC10.
The zener on the cell input would clamp the cell transient if it is due to operation so that less comes through the filter. It will also limit the voltage at VC10 during conneciton because once the diode conducts current will flow through the input resistor and diode to prevent excessive voltage build up on the capacitor. VC10 can still see the voltage of the diode plus the voltage across the resistor. If your zener diode limiter can be fixtured to the board during cell connection and removed after, it could be placed on VC10 for a more restrictive limit, then removed before operation.