Exchange only lines, which run direct to the MDF in the exchange without going through a PCP (cabinet) fall into two broad categories, which can be summarised as "near exchange" and "other". Those I classify as "near exchange" are the sort you describe - lines so close to the exchange that there was felt to be little point establishing a PCP as it was just as easy to run them direct to the MDF. The "other" lines are a much more diverse group, which include lines in areas so sparsely populated that a PCP wouldn't cover enough lines to be worth establishing.
There's four solutions to provide FTTx on the "near exchange" lines:
- establish a PCP near the exchange and re-route the lines through it
- install FTTP
- use future FTTDP (fibre to the distribution point) technology as a sort of mini FTTC cabinet for small groups of exchange-only lines
- run VDSL2 from the exchange
The problem is that there's a cost attached to all of these, whilst customers on near exchange EO lines are potentially the least likely to take FTTx products because they already enjoy high ADSL2+ speeds.
Moving the lines to a PCP has been done in some areas, sometimes using BDUK money. In your case it might be possible to rearrange the network to run some of the EO lines via PCP 13, making it worthwhile to deploy FTTC at that PCP.
FTTP roll-out is very slow nationwide at the moment.
FTTDP is under development at present, in the hope that running fibre even deeper into the network will provide higher speeds than FTTC without the costs of full FTTP, but in a way that can later be upgraded to FTTP.
I'd guess it is likely to be more expensive to use FTTDP for near exchange EO lines than establishing a conventional PCP and FTTC twin. However, there may be some scenarios where using FTTDP for EO lines is cost-effective, especially if the attempts to produce line powered FTTDP solutions come to fruition, removing the cost of a mains feed to the street equipment.
VDSL2 from the exchange is essentially prohibited by the Analogue Network Frequency Plan, because of the near-end crosstalk problems that would arise with originating ADSL and VDSL2 at the same location. It's possible that vectoring may allow the development of VDSL2 frequency plans and power masks suitable for VDSL2 from the exchange, but I think this unlikely. Vectoring may mitigate ADSL interference with VDSL2 lines, but is not likely to do anything for VDSL2 interference with ADSL lines.
VDSL2 from the exchange might only come about if, at a future time, ADSL from the exchange is withdrawn, either to be replaced by ADSL from the cabinet for those customers remaining on ADSL (which potentially has similar crosstalk issues to VDSL2 from the exchange, despite the power being lower), or because all remaining customers can switch to VDSL2.
From an equipment standpoint, discontinuing ADSL becomes easier over time. New consumer equipment will increasingly be ADSL/VDSL2 multi-standard, whilst modern MSANs and DSLAMs at most need a different line card for VDSL2 or ADSL. I have heard mention of at least one ADSL/VDSL2 multi-standard line card, but if I recall correctly it didn't support VDSL2 vectoring and had a lower port density than single-standard cards, both of which are serious drawbacks.
A potentially serious problem with VDSL2 for all is long lines. Others are more qualified to comment than I am, but my understanding is that VDSL2 typically performs less welle on long lines than ADSL. If VDSL2 cannot be made to perform as well as ADSL on long lines, moving all ADSL customers to VDSL2 is clearly a non-starter.