Accuracy of diffusion weighted MRI in detecting residual tumor in patients conservatively treated for cervical carcinoma.

High stage cancers of the uterine cervix are treated with curative intention,
by radiotherapy, in combination with neoadjuvant or concomitant chemotherapy or
with hyperthermia [1]. After completion of this treatment, 10 to 20 percent of
the patients still have residual tumor in the cervix [2]. Indeed, a recent
article from Vincens et al indicated histopathological evidence of residual
cervix tumor in no less than 36.5 % of the cases [3].
Early detection of residual tumor will likely enhance patient survival since a
salvage hysterectomy can be performed [2]. A retrospective study found that
salvage surgery cured 38% of patients with central residual tumor [2], and the
authors debated whether more extensive surgery in selected patients might be
beneficial and improve this number. Residual tumors detected at a later stage
(i.e. when clinically manifest) often need more elaborative therapies with
far-going quality of life effects, like a total exenteration (exenteration of
bladder, uterus, vagina and rectum). However, these patients are often directed
for palliative procedures. Not only patient perspective but also the financial
cost are significantly augmented hereby.

It is unclear what procedures are best used to assess tumor status after
treatment. Clinical investigation is hampered by the fibrosis resulting from
radiation therapy, and biopsy has a high rate of false negatives. In a study by
Nijhuis et al, clinical examination with biopsies had a sensitivity of slightly
more than 50% [2].
MRI is superior over CT scanning for the evaluation of the cervical region and
is now used as a standard technique used for pretreatment evaluation of the
cervical region[5]. Very recently guidelines of the European Society of
Urogenital Radiology are published that encompass clinical examination and post
treatment MR evaluation [6]. MR criteria for a complete response include;
- no lesions seen in the cervix or in the adjacent anatomic areas
- homogeneous hypointense cervical stroma
- Homogeneous and delayed intravenous contrast medium uptake of the cervix
after IV injection.

Recently, MRI with diffusion weighted imaging (DWI) was introduced with
promising results in detecting tumor in comparison with the standard techniques
used in MRI [7-9]. DWI is an imaging sequence that can be used in combination
with the standard sequences, and it takes only several minutes additional time.
The local tissue diffusion of cervical cancer proved to be significantly
different from normal cervical tissue, as shown by McVeigh et al [8]. The most
important improvement compared to standard MRI is the better differentiation
between tumor and fibrosis and edema, since only the first appears bright on
MRI-DWI. It is unlikely that radiation therapy influences the applicability of
DWI-MRI. In rectal tumors, this technique is even used during radiation therapy
[10]. Indeed, shortly after finishing radiotherapy, edema is present in the
tissues, and this improves the discriminating properties of particularly
MRI-DWI; in DWI edema is suppressed with better delineation of the high
intensity tumor; therefore even small residual tumor areas might be detected.
This edema might explain the relatively poor performance of standard MRI in the
detection of small tumor residues, as shown by Hatano et al., since both tumor
and edema are hyperintense on T2-weighted images [11].
According to the recent guidelines of the European Society of Urogenital
Radiology it is recommended to perform a MRI after radiation therapy in order
to evaluate tumor response. Response evaluation with MRi is then based on
conventional MRI sequences with the use of intravenous contrast. DWI findings
will not influence the decisions for further diagnosis and treatment since no
validation of this technique has been performed until now.We plan to
investigate the additional use of DWI in the detection of residual tumor in
patients conservatively treated for cervical cancer, above respons evaluation
by clinical examination combined with conventional MRI with the use of
intravenous contrast.
DWI findings will be analysed a year after treatment and will not influence the
decisions for further diagnosis and treatment.
In order to have a sufficiently large patient pool for meaningful analysis, we
intend to cooperate with another large tertiary center in Europe, namely
University Hospitals of Leuven in Belgium. This hospital offers a large patient
cohort, and also vast experience with diffusion imaging.
In a prospective cohort study, the accuracy of MRI-DWI in detecting residual
cervical cancer will be assessed. When MRI- DWI proves accurate in detecting
residual disease, MRI- DWI can be used as routine follow-up examination. This
will prevent unnecessary treatment delay, and improve the selection of patients
with potentially salvageable disease.

To determine the accuracy of MRI with DWI in detecting residual disease in
patients with uterine cervical cancer treated with (chemo)radiotherapy.
To determine the additional value of using DWI to a standard MRI protocol.
To evaluate in the inter-reader variability in reading DWI images in the
post-treatment setting

In the Universital hospitals of Leiden, Amsterdam en Utrecht, the second MRI
after radiation therapy will be performed as additional investigation. In this
situation, the findings of the DWI sequence will not influence the standard
care, whilst important findings during the second MRI may alter treatment. This
will be based on the conventional sequences and not the DWI sequence.

In the Universital hospitals of Rotterdam and Utrecht, the second MRI after
radiation therapy will be performed as part of the standard care. Also in this
situation , the findings of the DWI sequence will not influence the standard
care, whilst important findings during the second MRI may alter treatment. This
will be based on the conventional sequences and not the DWI sequence..

Prospective multicenter observational study

no burden, no risks, no benefit