Discussion
This case is unique because the cervical cytology changed in a short
time. The change in cervical and vaginal stump cytology without
treatment is rare, so we speculate that this change was associated with
allo-SCT.
GVHD is among the most potentially serious complications of allo-SCT,
but patients who develop GVHD also have a low incidence of leukemia
recurrence,1 suggesting a GVT effect due to the donor
lymphocytes in association with GVHD.
Eibl et al. reported a breast cancer patient with liver and bone
metastases who underwent allo-SCT and had complete resolution of the
liver metastases on CT 27 days later.2 On the same
day, the patient developed acute GVHD of the skin. Considering this
clinical course, Eibl et al. attributed the resolution of the metastases
to a GVT effect. Child et al. performed allo-SCT in 19 patients with
refractory metastatic renal-cell carcinoma, of whom 3 had complete
response and 7 had partial response. 3 They noted that
regression of metastases occurred at a median of 129 days after
transplantation, often following withdrawal of cyclosporine and the
establishment of complete donor T cell chimerism. Cyclosporine
suppresses the increase of T cells and the production of cytokines. This
agrees with the previous study suggesting that the graft-versus-leukemia
and GVT effect is caused by donor T cells.4
There are many reports of the GVT effect in patients with solid tumors.
To our knowledge, however, this is the first report showing a GVT effect
for only abnormal cytology. Some studies have reported that the cervical
cytology after allo-SCT was affected by immune status.
Shanis et al. studied 82 women who underwent allo-SCT and found the
cumulative incidence rate of any genital HPV infection at 20 years after
transplantation to be 40.1%.5 Women who developed
extensive and genital chronic GVHD had a cumulative HPV infection rate
of 67.1%, compared with 18.4% in women without chronic GVHD. They also
reported that the cumulative rate of severe dysplasia reached 19% at 20
years after transplantation, and the rate was significantly different
between women with and without chronic GVHD (41.2% versus 2.5%). In
contrast, there was no association between acute GVHD and the rate of
HPV infection. Shanis et al. suggested that, since sexual activity after
allo-SCT is generally decreased, the increased HPV infection rate was
due to reactivation of HPV rather than new infection. Moreover, viral
reactivation may have been influenced by chronic GVHD and/or
immunosuppression. Similarly, Bipin et al. reported that genital HPV
disease occurred in one-third of 38 patients.6 All of
them were long-term survivors of allo-SCT on prolonged immunosuppressive
therapy. Bipin et al. suggested that prolonged systemic
immunosuppressive treatment for chronic GVHD was associated with a
higher risk of developing genital HPV disease.
Our patient may have already been infected with HPV before she underwent
the first allo-SCT considering her history of sexually transmitted
infection. The abnormal cervical cytology may represent HPV reactivation
due to prolonged immunosuppression.
We suggest that our patient developed a GVT effect associated with GVHD,
which resulted in transient improvement of the cervical cytology after
the second allo-SCT. This GVT effect may have been suppressed later by
immunosuppressive therapy, resulting in worsening of her cervical
cytology. As shown in Figure 10, the cervical cytology changed from NILM
to HSIL. When GVHD recurred and the immunosuppressant dose was
increased, the cervical cytology improved. When the immunosuppressant
dose was again decreased because of improvement of GVHD, the cervical
cytology worsened. Although we speculate that the cervical cytology was
affected by a GVT effect mediated by the immunosuppressant dose, there
is no direct evidence that the cervical cytology improved because of a
GVT effect. However, transformation of the cervical cytology from NILM
to HSIL without treatment is rare. Although the possibility of sampling
error cannot be excluded, if it were to occur, then there would be no
correlation between cytology, GVHD and the immunosuppressant dose.
The cervical cytology of patients after allo-SCT is likely to change
depending on their immune status. Regular gynecologic follow-up is
needed even after improvement of the cervical cytology to monitor for
changes.