Regarding PreRender, how can I determine the appropriate off-screen area for 3D Collapsed Layers (or text and shapes)?

The plugin I'm developing uses PF_RenderOutputFlag_RETURNS_EXTRA_PIXELS.
What I want to do is render collapsed layers that are outside the composition's screen area within the screen.

When using Collapse, if I use the default `extra->input->output_request` without modifying it, information outside the composition screen (+20%) gets truncated.

In 2D, I can use the following code to obtain the range that isn't truncated.

Code

PF_CheckoutResult vInputDt;
PF_RenderRequest vReq = extra->input->output_request;

//Get default data.
ERR(extra->cb->checkout_layer(vFxRef, vLyrPrmIdx, vImgIDTp, &vReq, vIntTm, vTmStep, vTmSc, &vInputDt));

//Get main data from max rect.
PF_Rect vOglMaxRect = vInputDt.max_result_rect;
vReq.rect = vOglMaxRect;
ERR(extra->cb->checkout_layer(vFxRef, vLyrPrmIdx, vImgID, &vReq, vIntTm, vTmStep, vTmSc, &vInputDt));

However, it doesn't work in 3D.
Even if I expand the extra->input->output_request further,
I still only get the range of the composition screen (+20%).

Is there a better way to do this?
Or is it a feature of After Effects that it doesn’t pass information from 3D Collapsed Layers located outside the screen?
The same behavior can be observed with the standard “Transform” effect.