Phase II Study of Recombinant Interleukin la and Etoposide ......1722 IL-la and Etoposide in Relapsed Osteosarcoma entry. Table I Patient characteristics Metastasis Zubrod Pathological
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Vol. 3, 1 721-1729, October /997 Clinical Cancer Research 1721
3 The abbreviations used are: OS, osteosarcoma; IL, interleukin: PR,
partial response; TNF-a, tumor necrosis factor a; NO, nitric oxide.
Phase II Study of Recombinant Interleukin la and Etoposide in
Patients with Relapsed Osteosarcoma 1
Laura L. Worth, Norman Jaffe,
Robert S. Benjamin, Nicholas E. Papadopoulos,
Shreyaskar Patel, A. Kevin Raymond,
Shu-Fang Jia, Carlos Rodriguez, Jacalyn Gano,
Mary Ann Gianan, and Eugenie S. Kleinerman2
Departments of Pediatrics IL. L. W., N. J., E. S. K.I, Cell Biology
[S-F. J., C. R., J. G., M. A. G., E. S. K.], Pathology IA. K. RI. andMedical Oncology [R. S. B., N. E. P. S. P.1, The University of Texas
M. D. Anderson Cancer Center, Houston, Texas 77030
ABSTRACT
A Phase II trial using interleukin hi (IL-la) and eto-poside for patients with relapsed osteosarcoma (OS) was
undertaken to assess the feasibility and tolerability of corn-bination therapy with biotherapy and chemotherapy. Nine
patients with histologically proven relapsed OS were treated
with IL-la immediately followed by etoposide daily for 5
days every 3 weeks. Surgical resection of lung metastasis or
peripheral tumor was performed after two or three cycles.We observed three partial responses; disease was stable inanother case. One case could not be evaluated. The sideeffects associated with combination therapy were as pre-
dicted from known side effects of the individual agents;
however, more profound neutropenia was observed. Four
patients exhibited clinical signs of capillary leak syndrome,
i.e., hypotension, edema, and weight gain. The etiology of the
capillary leak was unclear, because serum IL-la, IL-2, tu-mor necrosis factor, and nitric oxide levels could not be usedto predict which patients would develop capillary leak. His-tological analysis of tumor specimens obtained after two or
more courses of therapy showed changes consistent with a
response to a biological response modifier: peripheral fibro-
sis surrounded the metastasis with infiltration of chronic
and acute inflammatory cells. Because the response of Fe-lapsed OS to any type of salvage regimen has been poor, we
interpret the clinical response of this therapy as good. How-ever, the significant side effects associated with this therapy
must also be taken into consideration before deciding to use
Received 4/30/97: revised 7/1/97: accepted 7/8/97.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby marked
advertisement in accordance with 18 U.S.C. Section 1734 solely to
indicate this fact.
I Supported in part by Grants CA-l7772, CA-42992, and CA-66632.
Also supported by a grant from the Gillson-Longenbaugh Foundation
(to E. S. K.).2 To whom requests for reprints should be addressed. at Department of
Cell Biology, Box 173, The University ofTexas M. D. Anderson CancerCenter, 1515 Holcombe Boulevard, Houston, TX 77030. Phone: (713)
792-81 10: Fax: (713) 792-8747.
this combination therapy. It is unfortunate that the study
was stopped early due to halted production of IL-la. If this
agent is again manufactured for clinical use, we conclude
that additional evaluation in patients with relapsed OS iswarranted.
INTRODUCTION
The majority of patients with OS� have subclinical pulmo-
nary micrometastases at the time of initial presentation and
diagnosis. Surgical resection of the primary tumor coupled with
intensive adjuvant chemotherapy has improved the metastasis-
free survival rate from less than 20% at 2 years to 60% at S years
( 1-3). However, 30-40% of these patients still die of pulmo-
nary metastases that are resistant to conventional therapy (2-4).
The majority of patients experience a relapse within the first 2
years, either while receiving adjuvant chemotherapy or shortly
after the completion of chemotherapy. The prognosis for these
patients is poor. Although the metastases can be removed by
surgery, 70-85% will recur within 1 year (2-4). Several che-
motherapeutic agents, different from those used in the initial
therapy, are being studied in an attempt to develop more effec-
tive therapy.
One agent that has been studied is etoposide, an intracel-
lular poison that acts by stabilizing the cleavable complex
between topoisomerase II and DNA. The drug-topoisomerase II
DNA complex formation initiates a series of intracellular events
culminating in the death of the cell (5). The experience with
etoposide used as a single agent in relapsed OS is limited. A
Pediatric Oncology Group study (6) analyzed the effect of
etoposide when ISO mg/m2/day were given IV for 3 days every
3 weeks. Of the 1 1 patients studied, I PR and 1 minor response
were observed. Consequently, etoposide as a single agent is not
considered active against relapsed OS.
IL-l is primarily a monocyte/macrophage product pro-
duced in response to infection, injury, or immune challenge.
Among its numerous biological activities are inducing other
cytokines [e.g., IL-2, IL-6, colony-stimulating factors, TNF, and
intracellular adhesion molecules (7-9)] and activating T and B
lymphocytes (10-12). It also enhances the ability of lympho-
cytes to bind and kill tumor cells (13). In melanoma cells, a
synergistic interaction between IL- I a and etoposide has been
reported (14).
Recent studies (IS, 16) demonstrated that IL-la enhanced
the in vitro cytotoxicity of etoposide. In three different human
OS cell lines with relative resistance to etoposide, use of IL- 1a
increased the cytostatic activity of etoposide from 9% to 70%.
The synergistic activity was only demonstrated when IL-la
incubation preceded or coincided with etoposide. When etopo-
Plasma was fractionated, separated, frozen at-70#{176}C,and subsequently assayed for IL-2. Notethat plasma samples from patient 2 were col-lected during an episode of capillary leak.
0U
z
30
20
10
0
Fig. 4 Plasma levels of NO following the ad-ministration of IL-la and etoposide. Blood sam-
ples were collected from patients 2-9 at various
times after etoposide and IL-la administration.
Plasma was separated, frozen at - 70#{176}C,and sub-sequently assayed for NO. Note that plasma sam-
ples from patient 2 were collected during an epi-
sode of capillary leak.
Time (hours)
plasma TNF-a levels were seen in five of eight patients follow-
ing combination therapy.
The etiology underlying the capillary leak syndrome seen
in four patients was also examined. IL-2 and NO have been
associated with hypotension and peripheral capillary leak syn-
drome. Therefore, patient plasma samples were also examined
for levels of IL-2 and NO. Figs. 3 and 4 illustrate that there was
no correlation between plasma IL-2 or NO levels and the de-
velopment of hypotension and edema following IL-la and
etoposide therapy. Significant increases in plasma IL-2 were
only seen in patients 3 and 5. Capillary leak syndrome did not
develop in patient 3 but was apparent in patients 2, 4, and 5.
Fig. S Chest radiographs ofpatient 1 . Studies were obtainedprior to (left panel) and after(right panel) two courses of eto-poside and IL- 1a. An increase is
seen in the metastasis in theright upper lobe. Decreases areseen in the size of the left mid-dle and upper lobe metastases.No new disease is seen.
Clinical Response. As mentioned earlier, an adverse re-
action after the initial dose of IL-la and etoposide prevented
one patient (patient 2) from being included in our evaluation. Of
the remaining eight patients, disease in three patients responded
partially to therapy. These patients received five to nine courses,
and their PR lasted for 6-8 months.
Patient 1 presented with bilateral pulmonary metastases.
After the patient received two courses of combination therapy,
a chest X-ray showed enlargement of the pulmonary mass on the
right and decreases in the size of several lesions on the left (Fig.
5). Initially, this case was classified as a mixed response. A
thoracotomy was performed after the second course of treatment
because of the confusing clinical picture and what was pre-
sumed to be an enlarging mass in the right lung. The operation
revealed a cavitated spherical cyst filled with fluid. Histologi-
cally the cystic mass (Fig. 6) was composed of cellular debris,
inflammatory cells, and granulation tissue, with small pockets of
viable tumor cells only at the peripheral rim of the cyst. There-
fore, we concluded that the radiographic expansion of the mass
was not due to tumor growth but rather was secondary to
collection of fluid and inflammatory elements within the cystic
lesion. In view of the shrinking lesions in the left lung and the
necrotic tumor cells in the right lesion, we classified the case as
a PR. This patient received two additional courses of combined
therapy, and the lesions in the left lung were surgically resected.
Here too, the histological examination revealed intralesional
hemorrhage, fragments of dishesive tumor with some viable
tumor cells, and moderate infiltration with both acute and
chronic inflammatory cells. The findings are not typical of those
seen with chemotherapy in which the metastatic lesion has
central necrosis surrounded by a rim of viable tumor cells
growing into compressed normal lung tissue ( I 7).
The second patient (patient 7) to achieve a PR underwent
surgical resection of nodules in the right lung after three courses
oftherapy. On the basis of results from radiographic studies, this
patient’s disease was classified as stable. Histological examina-
tion of the tumor mass showed greater than 80% necrosis. The
majority of what were interpreted initially as “viable” cells were
shown by immunohistochemical studies to be inflammatory in
origin. Consequently, the case was also classified as a PR based
on histological examination.
The final patient in the PR category (patient 8) also had
bilateral pulmonary metastases; one lesion was almost corn-
pletely resolved, and the others decreased in size with treatment.
This patient had a single surgical excision of the pulmonary
metastases after three courses of treatment. The remaining mass
showed liquefaction and necrosis. Some cell dropout suggestive
of a response to a therapeutic agent was noted. However, there
was cautery damage to the tissue, which made the interpretation
less conclusive.
Although mixed response is not a traditional category, parts
of tumors in two patients had dramatic responses. Patient 6 had
OS of the hard palate. At this patient’s entry into the study, the
tumor mass was pushing the tongue posteriorly, partially ob-
structing the aerodigestive tract and requiring tracheostomy and
gastrostomy. After three courses of chemotherapy, the intraoral
portion of the tumor had regressed sufficiently to allow the
patient to speak more clearly and swallow. On radiographic
studies, the tumor progressed anteriorly, compromising the optic
chiasm and resulting in progressive loss of vision in one eye.
The other patient in this category (patient 4) had metastatic
disease to the lung as well as to the humerus and surrounding
soft tissue. She underwent surgical resection after two courses of
IL- I a and etoposide to assess response to therapy; the decrease
in pulmonary disease was accompanied by increased disease
invasion of soft tissue around the humerus. Pathological speci-
mens of the pulmonary nodules obtained after two courses of
therapy contained more than 80% necrosis. The humeral in-
tramedullary disease had >90% necrosis; however, the ex-
tramedullary, soft tissue component only exhibited 60% necro-
sis. In this case too, one area of the tumor responded
dramatically to therapy, whereas in another area the response
was less impressive.
Patient 3, with previous rapidly progressive pulmonary
disease, exhibited stable disease when treated with IL-la and
OS lung metastasis removed at-ter therapy with etoposide andIL-la in patient 1 . A, residual
tumor (T between brackets) and
underlying fibrosis (F) form athin rim at the interface between
normal lung (NL) and central
cystic cavity (CC; H&E; mag-nification, X40). B, “residualtumor” is made up of a heterog-enous mixture of elements. Thetumor itself, with the accompa-nying osseous matrix (solid ar-
rows), is composed of appar-ently unaffected neoplastic cellsand cells with significant degen-erative cytological changes, aswell as cell dropout (hollow ar-
rowheads). There is a mixed in-
flammatory infiltrate scattered
diffusely throughout the lesion
(Infi), and it includes lympho-cytes, plasma cells, histiocytes
(some hemosiderin laden), and
polymorphonuclear cells (H&E;magnification, x 100).
Two patients had progressive disease. Brain metastases
developed in one patient (patient 5) after the second course of
chemotherapy. This patient was removed from the protocol.
Lung disease developed in the other patient (patient 9) after the
: � �
‘I�4 �
‘C’s_ . �
�
�.J
ja.. 4�#{149} .r
� � �1
.,±, 4�?
Q� �
�A �
‘.
-�. ‘�‘1’..��;- .F�f1w ‘ � � ,
,-. � S� , ‘ ‘�
� .� �
first course, but this patient received a second course, because
the disease had responded poorly to previous therapies; there
were essentially no other treatment modalities available. After
the second course, radiographic studies indicated some of the
pulmonary nodules responded to therapy with necrosis and
liquefaction, whereas other areas did not. Low pulmonary
reserve precluded this patient from being a candidate for
thoractomy.
DISCUSSION
Although we were unable to complete this Phase II trial
because IL-la became unavailable, we demonstrated that the
biotherapy-chemotherapy combination is feasible and tolerable.
The side effects associated with the study were predicted from
the known side effects of each agent. However, the profound
neutropenia was greater than that experienced when etoposide
was used as a single agent. Fever and neutropenia were signif-
icant side effects requiring hospitalization of the affected pa-
tients. Sixteen of the 35 courses were associated with fever,
neutropenia, and blood cultures negative for bacteria. Seven
courses (20%) were associated with positive blood cultures in
addition to neutropenia. The profound neutropenia may be ex-
plained by the higher median daily area under the curve and
lower clearance rates quantified in patients receiving combina-
tion IL-Ia and etoposide compared with those values reported
when etoposide is administered alone (18). Alternatively, the
synergistic interaction that caused increased tumor killing of
cells may also have caused increased killing of bone marrow
cells. Only one patient required reduction in etoposide dose (for
two courses); it was restored to 100% for the remainder of the
courses.
Four patients in five courses of therapy had symptoms
suggestive of peripheral capillary leak syndrome (hypotension,
edema, and weight gain). Serum levels ofthe proteins associated
with this syndrome (IL-l , IL-2, and TNF-a) and NO were not
helpful in predicting which patients would develop capillary
leak. The factors involved in the development of this syndrome
are still unclear.
A difficult problem in analyzing tumor response to a bio-
logical response modifier is measuring the clinical response to
the therapy. In situations in which cytokines are administered or
their release is stimulated, an inflammatory reaction can be
initiated. By conventional radiography, this inflammatory prod-
ess can appear as an enlarging tumor mass and be interpreted as
worsening disease. Consequently, evaluating the results of
chemotherapy plus biological response modifiers against meta-
static lung disease by use of a chest radiograph may not truly
represent what is happening on the cellular level. In several
patients, an inflammatory response was initiated that resulted in
fluid accumulation either as a cystic mass surrounding the
metastatic nodules or actually replacing the disease. If a lesion
is partially filled with fluid, an air-fluid level can be appreciated
and may differentiate such a mass from an enlarging metastasis.
When results of therapy were not clear, surgery was performed
without waiting for the third course of chemotherapy to be
completed. We correlated the histological characteristics of
these enlarging metastatic lesions with their appearance on
nuclear magnetic imaging scans to better interpret the effects of
therapy.
For most patients, the time needed to recover from thoracic
surgery was usually 2 weeks. Consequently, the time interval
between courses of etoposide and IL-la was 6 to 8 weeks. Two
patients had two surgical procedures between courses of chem-
otherapy (sequential thoracotomies and thoracotomy followed
by hemipelvectomy). For these patients, the time between
courses of chemotherapy when separated by two surgical pro-
cedures approached 3.5 months. During this interval without
therapy, it was not uncommon for the tumor that had been
responding to chemotherapy to begin to grow again.
Because the response of relapsed OS to any type of salvage
regimen has been poor, we interpreted the clinical response to
this therapy as good, with three PRs and one case of stable
disease in eight patients. All four patients had previously failed
a number of salvage regimens in addition to their front-line
chemotherapy. If the assessment of the response to therapy were
determined by more conventional methods, using only radio-
graphic evidence, than only one patient would be considered to
have had a PR. Judged by traditional criteria, this therapy would
have the same overall outcome as in the Phase II study evalu-
ating etoposide as a single agent (6). However, biological re-
sponse modifiers, such as IL-la, act by stimulating the immune
system, enhancing the cytotoxic activity of agents, and inhibit-
ing angiogenesis (19). Consequently, determining the response
rate using only radiographic evidence when there is a mass
effect from inflammation and hemorrhage could result in un-
derestimation of response and the possible elimination of a
potentially useful combination therapy. Under strict criteria, the
mixed response category is considered progressive disease.
However, it was a distinct category in this study because parts
of two patients’ tumors showed response to therapy. In the
patient with metastatic disease to the humerus that extended into
the adjacent soft tissue, the intramedullary disease showed
greater than 90% necrosis histologically. She had no therapy for
4 weeks before beginning IL-la and etoposide; thus this necro-
sis is unlikely to have been the result of prior therapy. The soft
tissue disease failed to respond to therapy. For the patient with
OS of the hard palate, tumor regression restored the ability to
speak more clearly and eat again; this was a welcome relief and
a dramatic response.
Halted production of IL-la caused this study to be stopped
prematurely. If this agent is again manufactured for clinical use,
we conclude that its further evaluation in patients with relapsed
OS is warranted. Although some side effects were significant
and need to be taken into consideration, the rate of disease
response in this study of combined biotherapy and chemother-
apy makes this approach promising.
REFERENCES
I . Jaffe, N. Chemotherapy in osteosarcoma: advances and controver-
sies. In: F. M. Muggia (ed), Experimental and Clinical Progress inCancer Chemotherapy, pp. 223-233. Boston: Martinus Nijhoff Publish-
ers, 1986.
2. Link, M. P., Goorin, A. M., Miser, A. W.. Green, A. A., Pratt, C. B.,Belasco, J. B., Pritchard, J., Malpas, J. S., Baker, A. R., Kirkpatrick,J. A., Ayala, A. G., Shuster, J. J., Abelson, U. T., Simone, J. V., andVietti, T. J. The effect of adjuvant chemotherapy on relapse-free sur-vival in patients with osteosarcoma of the extremity. N. Engl. J. Med.,314: 1600-1606, 1986.
3. Eilber, F., Giuliano, A., Eckardt, J., Patterson, K., Moseley, S., andGoodnight, J. Adjuvant chemotherapy for osteosarcoma: a randomized
prospective trial. J. Clin. Oncol., 5: 21-26, 1987.
4. Goorin, A. M., Perez-Atayde, A., Gebhardt, M., Anderson, J. W.,
Wilkinson, R. H., Delorey, M. J., Watts, H., Link, M., Jaffe, N., and
Frei, E. Weekly high-dose methotrexate and doxorubicin for osteosar-
coma: the Dana-Farber Cancer Institute/The Children’s Hospital-
Study III. J. Clin. Oncol., 5: 1 178-1 184, 1987.
5. Ross, W., Rowe, T., Glisson, B., Yalowich, J., and Liu, L. Role of
topoisomerase II in mediating epipodophyllotoxin-induced DNA cleav-
age. Cancer Res., 44: 5857-5860, 1984.
6. Kung, F., Hayes, F. A., Krischer, J., Mahoney, D., Jr., Leventhal, B.,Brodeur, G., Berry, D. H., Dubowy, R., and Toledano, S. Clinical trialof etoposide (VP-16) in children with recurrent malignant solid tumors:a Phase II study from the Pediatric Oncology Group. Invest. New Drugs,
6: 31-36, 1988.
7. Le, J., and Vilcek, J. Tumor necrosis factor and interleukin- 1 : cyto-
kines with multiple overlapping biological activities. Lab. Invest., 56:
234-248, 1987.
8. Ochoa, A. C., Gromo, G., Alter, B. J., Sondel, P. M., and Bach, F. H.
Long-term growth of lymphokine-activated killer (LAK) cells: role of
anti-CD3, beta-ILl, and interferon-gamma and -beta. J. Immunol., 138:
2728-2733, 1987.
9. Belardelli, F., Proietti, E., Ciolli, V., Sestili, P., Carpinelli, G., Divito,M., Ferretti, A., Woodrow, D., Boraschi, D., and Podo, F. Interleukin-143 induces tumor necrosis and early morphologic and metabolicchanges in transplantable mouse tumors: similarities with the anti-tumoreffect oftumor necrosis factor a or �3. Int. J. Cancer, 44: 1 16-123, 1989.
10. Farran, W., Mizel, S., and Farran, J. Participation of lymphocyte
activation factor (IL-l) in the induction of cytotoxic T-cell responses.
J. Immunol., 124: 1371-1377, 1980.
1 1. Dower, S. K., Kronheim, S. R., March, C. J., Conlon, P. J., Hopp,
T. P., Gillis, S., and Urdal, D. L. Detection and characterization of high
affinity plasma membrane receptors for human interleukin I . J. Exp.
Med., 162: 501-515, 1985.
12. Dinarello, C. A. Biology of interleukin 1. FASEB J.. 2: 108-1 15.
1988.
13. Oppenheim, J. J., Kovacs, E. J., and Matsushima, K. There is more
than one interleukin-l . Immunology, 7: 45-56, 1986.
14. Usui, N., Mimnaugh, E. G., and Sinha, B. K. Synergistic antitumor
activity of etoposide and human interleukin-la against human mela-
noma cells. J. Natl. Cancer Inst., 81: 1904-1909, 1989.
15. Jia, S-F., Zwelling, L. A., and Kleinerman, E. S. The synergisticeffect of IL-l and VP-l6 against osteosarcoma cells. Proc. Am. Assoc.
Cancer Res., 32: 227, 1991.
16. Jia, S-F., Zwelling, L. A., McWatters, A., An, T., and Kleiner-
man, E. S. Interleukin-l-a increases the cytotoxic activity of etopo-
side against human osteosarcoma cells. J. Exp. Ther. Oncol., in press.
1997.
17. Kleinerman, E. S., Raymond, A. K., Bucana, C. D., Jaffe, N.,
Harris, M. B., Krakoff, E. H., Benjamin, R., and Fidler, I. J. Unique
histological changes in lung metastases of osteosarcoma patients fol-
lowing therapy with liposomal muramyl tripeptide (CGP l9835A lipid).
Cancer Immunol. Immunother., 34: 21 1-220. 1992.
18. Soefje, S. A., Madden, T., Herzog. C., Jaffe, N., Gano, J. B., and
Kleinerman, E. S. The effect of IL- Ia on etoposide pharmacokinetics in
patients with relapsed metastatic osteosarcoma. Pharmacotherapy, 15:
375, 1995.
19. Potvin, F., Petitcler, E., Marceau, F.. and Poubelle, P. E. Mech-anism of action of antimalarials in inflammation: induction of
apoptosis in human endothelial cells. J. Immunol.. /58: 1872-1879,
1997;3:1721-1729. Clin Cancer Res L L Worth, N Jaffe, R S Benjamin, et al. in patients with relapsed osteosarcoma.Phase II study of recombinant interleukin 1alpha and etoposide