TARGETED AND UNTARGETED FUSION PROTEINS: CURRENT APPROACHES TO CANCER IMMUNOTHERAPY - Fusion Protein Technologies for Biopharmaceuticals: Applications and Challenges - page 298

initiation of therapy, however, showed no difference
between na ıve and remission mice. To determine if regen-
erated CD4 þ CD25 þ T cells in tumor-draining lymph
nodes correlated with the ability of tumor to take in these
mice, the percent of T reg cellswasmeasuredineachof
these groups of mice. The data showed that CD4

TABLE 19.4 Combination Cytokine or Chemokine Fusion
Protein Immunotherapy and CD4þ T-cell Depletion in the
Treatment of COLON 26 Carcinoma

Tumor
Reduction c (%) Immunotherapy a T-cell Subset Depletion b

þ CD25 þ

chTNT-3 (control) CD4 þ depletion 33 (0)

T cells in na ıve mice constituted 9.7% of total T cells,

while in 2-3 month tumor-regressed mice they constituted
only5%oftotalTcellsandin5-6monthstumor-regressed
T reg cells constituted 13% of total T cells. This increase in
T reg cells may explain the failure to reject tumor seen at 6
months. Additionally, these experiments suggest that com-
bination LEC/chTNT-3 treatment and CD25

€

LEC/chTNT-3 CD4 þ depletion 100 (60)

chTNT-3/IL-2 CD4 þ depletion 64 (38)

chTNT-3/IFN- g CD4 þ depletion 33 (32)

chTNT-3/TNF- a CD4 þ depletion 33 (10)

Values in parentheses refer to group of animals which are not treated with
CD4 þ depletion.
a Antibodies and fusion proteins (20 m g/dose) were injected i.v. for five
consecutive days after tumors reached 0.5 cm in diameter.
b CD4 þ depletion (0.5mg/dose of GK1.5) was performed i.p. 1 day after
tumor implantation and repeated every 5 days.
c

depletion
does in fact produce long-acting memory cells capable of
preventing re-engraftment of the same tumor, so long as
T reg regeneration does not suppress the memory response.
To show that targeted LEC is critical to the above results,
similar combination studies were performed with chTNT-
3/cytokine fusion proteins consisting of human IL-2, murine
IFN- g , and murine TNF- a using identical treatment regi-
mens. The results of these studies presented in Table 19.4,
showed only modest improvement indicating that combina-
tion therapy with anti-CD25

þ

Percentage tumor reduction on day 19 after treatment.

in the peripheral blood. As expected, depletion of CD8 and

NK cells negated the antitumor activity of LEC/chTNT-3
treatment, indicating that these lymphocyte subpopulations
are critical to the function of the fusion protein (Table 19.4).
By contrast, mice receiving CD4 þ depletion regardless of
the time of administration in combination with LEC/chTNT-
3 therapy demonstrated 100% regression of tumor. These
unexpected results were highly significant and turned our
attention to the necessity of deleting or suppressing immu-
noregulatory cells and mechanisms.
To probe this observation further, only the CD4

þ

antisera requires LEC not
cytokine localization to tumor in order to produce complete
regression.

þ

19.5.2 CD137L and GITRL Fusion Proteins
in Combination with T reg Depletion
As described above (Figure 19.3), mice treated with
B7.1/NHS76 alone showed a 35-55% reduction in tumor
volume. To determine if T reg deletion could enhance the
effects of B7.1 fusion proteins, additional tumor-bearing
mice were treated with combination therapy using PC61 and
B7.1/NHS76 [17]. Indeed, combination therapy produced
complete regression of established lung tumors (MAD109)
and was associated with increased effector T cell tumor
infiltration (Table 19.5). When soluble B7.1-Fc was used in
combination with T reg depletion PC61 in MAD109-bearing
mice, all mice also went into complete remission and
remained so for the observation period of 6 months.
Rechallenge experiments performed with these mice
showed that they had immunological memory and were
able to reject subsequent tumor implants.
Using knockout mice, it was determined that the anti-
tumor activity of B7.1-Fc immunotherapy was dependent
both on IFN- g and perforin, but not on IL-4, as demonstrated
by the failure of IFN- g and perforin knockout mice to reject
tumors after treatment (Table 19.5). In these studies, anti-IL-
4 was found to have no tumor suppressive activity. By these
experiments, it appears that perforin may not be the exclu-
sive method of tumor cell killing since the activity of the
B7.1-Fc was not entirely eliminated as seen in the IFN- g
knock-out experiments. In summary, these data show that

þ
T reg subpopulation was depleted in repeat experiments.
Since T reg represents

þ CD25

population in
both mice and humans, depletion of this small subpopulation
using the rat antimouse CD25

10% of the CD4 þ

mAb PC61 would be highly
instructive. Depletion of T reg cells in combination with
LEC/chTNT-3 did indeed produce complete remissions in
COLON 26 tumor-bearing BALB/c mice, suggesting that it
is this small population of CD25

þ

cells that is responsible
minimizing the effects of LEC/chTNT-3. Treating with
PC61 alone or in combination with control chTNT-3, also
produced significant tumor regressions, but never a complete
response [15].
Rechallenge experiments demonstrated that combining
LEC/chTNT-3 treatment with CD25

þ

þ

depletion produces
long-acting memory cells. Tumor-free mice from the
experiments described above were again injected sub-
cutaneously with COLON 26 and MAD109 tumor cells
in their contralateral flanks, alongside na ıve mice injected

with each tumor cell line. Two weeks after tumor implan-
tation, all na

€

ıve mice had both tumors growing on their
flanks, while the tumor-remission mice (now 2 months
after initial tumor implantation) had only one tumor
(MAD109) growing. Interestingly, a second series of
rechallenge experiments performed 6 months after the

€