Erythropoietin and the Management of Anemia in Patients With Lung
Cancer
Jeffrey Crawford, MD, and Susan Blackwell, PA-C, MHS
The onset of anemia is a confounding factor that will affect the
majority of patients
with lung cancer. In an attempt to manage or prevent anemia in this population,
three trials have incorporated erythropoietin in the treatment schedules. The
results
of erythropoietin use for cancer patients with anemia have been encouraging.
Introduction
The majority of the 170,000 patients diagnosed with lung cancer in
the United States each year will develop anemia at some point, as a factor of the disease
and/or the result of treatment. Management of this complication in the past has been
limited to either excluding other causes of anemia in this population or providing red
blood cell transfusions for severely anemic patients. Three trials investigating the use
of recombinant erythropoietin (epoietin alfa) to treat or prevent anemia in cancer
patients, with an emphasis on patients with lung cancer, are reviewed.
The general indications for erythropoietin (EPO) use have expanded
greatly over the past decade, from anemia in chronic renal failure dialysis patients
initially in 1987, to predialysis and the HIV-related population in 1990, and to
chemotherapy-related anemia in 1993. In 1996, recombinant EPO was also approved for anemia
due to noncardiac surgery.
Three studies -- two completed and one ongoing -- are systematically
evaluating the effects of EPO on anemia in lung cancer patients. While many questions are
yet to be resolved, each of these trials will likely contribute additional information
concerning the clinical benefit of EPO in treating or preventing treatment or
disease-related anemia.
Registration Trial for Epoietin Alfa
In the original registration trial investigating recombinant EPO in
cancer chemotherapy-related anemia, patients were randomized to receive either placebo or
epoietin alfa.1,2 Both groups were anemic at presentation. The placebo group
continued to be anemic over the 12 weeks of therapy. By contrast, the EPO group had a
significant improvement of approximately 6 hematocrit points over this time period. This
improvement was seen in both the platinum-based and nonplatinum-based chemotherapy groups
who received EPO. In addition, a similar benefit was seen in the anemic nonchemotherapy
population receiving EPO, but this did not reach statistical significance, possibly
because treatment was continued for only eight weeks. The fact that differences were seen
in all three groups is an impressive outcome, given the cross-sectional nature of the
study. However, the positive results of this trial were difficult for community
oncologists to accept and incorporate in their practices because they represented a change
in the standard of care, and physicians were uncertain which patients would be most likely
to benefit.
Chemotherapy-Related Anemia: Community-Based Phase IV Trial for
Epoietin Alfa
This situation led to the development of a large, community-based,
phase IV open-label trial with 2,030 patients evaluated.3 Of the enrolled
patients, the lung cancer population was the largest single group of solid-tumor patients
(22%). All patients were treated with epoietin alfa at an initial dose of 150 U/kg SQ
three times per week over a four-month period.
Endpoints of this broad-based community trial were transfusion
requirement, hemoglobin level, and overall quality of life. A total of 1,475 patients
(approximately 75% of the population) were evaluable for quality-of-life endpoints, which
included energy level, daily activities, and overall quality of life.
Study Results
Patients demonstrated significant improvements in energy level,
daily activities, and overall quality of life. Approximately 250 patients in the group of
1,500 did not achieve an improved hemoglobin level. Another 500 achieved an improvement of
between 0 to 2 g/dL, and another 500 between 2 to 4 g/dL. More than 200 reached a
improvement of greater than 4 g/dL. The majority of the patients in this trial had an
improvement of 2 g/dL or more over this four-month period. Improvement in quality-of-life
measures increased linearly with the rise in hemoglobin, particularly between 8 to 12
g/dL. There was a continued improvement in the measures above a hemoglobin level of 12
g/dL, but the rate of increase was less. Using these parameters, this study suggests that
obtaining and maintaining a hemoglobin level of 12 g/dL or greater is optimal.
The lung cancer subpopulation included 438 patients, with an
approximately equal number of men and women. The group consisted of an older population in
general (Table 1). The mean hemoglobin level for lung cancer patients was 9 g, and
approximately one third had received transfusions. Median pretreatment serum EPO levels
were 65. Both carboplatin and cisplatin were commonly included in the treatment programs,
with a smaller group receiving a nonplatinum-based chemotherapy (Table 2). Approximately
two thirds of the overall patients became transfusion- independent. Of the majority who
were transfusion-independent at enrollment, only a smaller fraction became
transfusion-dependent on treatment.
Table 1. -- Lung Cancer Population: Patient Demographics and Baseline Characteristics
|
| Total patients |
|
438 |
| |
Men |
230 |
| |
Women |
208 |
| Mean age (years) |
|
64.5 |
| |
| Mean hemoglobin (g/dL) |
|
9.3 |
| |
| Patients transfused (%)* |
|
35.4 |
| |
| Mean units transfused/patient/month** |
|
0.57 |
| |
| Endogenous serum erythropoietin level (optional,
n=161) |
| |
| |
Mean (mU/mL) |
106 |
| |
Median (mU/mL) |
65 |
| * Within 4 months prior to study |
| ** Within 1 month prior to study |
| |
| From Clinical Outcomes: Lung Cancer. Educational product
flier: Procrit (Epoetin Alfa). Raritan, NJ: Ortho Biotech Inc; 1996. Reprinted with
permission. |
Table 2. -- Lung Cancer Subpopulation: Chemotherapeutic Agents (N=438) |
| Agent |
Number |
| |
| Cisplatin* |
|
189 |
| |
| Carboplatin* |
|
109 |
| |
| Cisplatin/carboplatin |
|
1 |
| |
| Nonplatinum (total) |
|
87 |
| |
| |
Doxorubicin |
30 |
| |
| |
Taxol** |
10 |
| |
| |
Unspecified |
47 |
| |
| * Patients may have received other agents. |
| ** Taxol is a registered trademark of the
Bristol-Myers Squibb Co. |
| |
| From Clinical Outcomes: Lung Cancer.
Educational product flier: Procrit (Epoetin Alfa). Raritan, NJ: Ortho Biotech Inc; 1996.
Reprinted with permission. |
Mean hemoglobin levels increased steadily from baseline through
the four months of therapy, while the proportion of patients requiring transfusion
decreased over this time (Fig 1). Because of their smoking history, lung cancer patients
are more likely than other types of cancer patients to have comorbid disease such as
cardiac disease or chronic obstructive pulmonary disease; thus, they become more
symptomatic from lesser degrees of anemia.
At the time of enrollment, 75% of patients were
transfusion-independent. With EPO treatment, 82% of these patients remained transfusion
independent over the course of the trial, while 18% developed transfusion dependence. In
the transfusion-dependent population, 68% became transfusion independent with EPO
treatment.
Adverse experiences from EPO treatment per se were virtually
nonexistent. Most of the reported adverse events (eg, disease progression, neutropenia,
leukopenia, thrombocytopenia) were likely effects from disease therapy and/or
chemotherapy. The study had a high dropout rate, with only approximately 50% of patients
completing the study. This rate reflects the nature of the study population and the
natural history of lung cancer patients. Disease progression, death, personal reasons,
intercurrent illness, and other adverse experiences were responsible for the dropouts.
Quality of Life in Lung Cancer Patients
When quality of life at baseline is compared with that at the
termination of this study, the data show a significant improvement in energy level,
activity, and overall quality of life for those achieving a 2 g/dL improvement or greater
in hemoglobin level. Evaluations were completed for 75% of the population. As noted in Fig
2, the parameters for quality of life improved in a stepwise fashion as a function of
increases in hemoglobin level.
Predictors of Overall Response
This and other studies show that an increase in hemoglobin level of
greater than 1 g/dL at four weeks (which was the case in approximately half of these
patients) predicts for further improvement in anemia. Eighty-one percent will ultimately
show at least a 2-g/dL improvement. This finding can be used as a predictor of overall
response. If the hemoglobin increase is less than 1 g/dL, only approximately a third
eventually demonstrate an improvement of 2 g/dL or more.
Conclusions From the Community-Based Study
The study concludes that an increase in hemoglobin levels in lung
cancer patients with anemia can be anticipated in most patients with the use of epoietin
alfa, and transfusion requirements were reduced. Improvements were achieved in energy,
activity, and overall quality of life, as measured by the linear analogue scale, and the
EPO treatment was well tolerated. The following guidelines have been proposed from the
data from the overall study for use in the clinical setting of cancer-related anemia.
• Evaluate for another cause of anemia, and if found, treat the cause.
• If no independent cause for anemia is found, start epoietin alfa at a dose of
10,000 U three times per
week subcutaneously.
• If patients do not respond, consider increasing the dose.
• If the patient's hemoglobin rises more than 1 g, continue with epoietin alfa
until normal hemoglobin is achieved.
• Decrease or discontinue the dose of epoietin alfa if hemoglobin becomes too high
(ie, more than 15 g/dL).
Preemptive Therapy With Epoietin Alfa
In the US registration trial of G-CSF (filgrastim), patients with
small-cell lung cancer received either placebo or G-CSF. G-CSF patients received
cyclophosphamide, doxorubicin, and etoposide in relatively high doses, followed by either
G-CSF support or placebo.4 Results showed that the number of patients who
developed anemia with hemoglobin levels of less than 8 was the same in both arms of the
study. Thus, anemia was not an effect of G-CSF per se but resulted from our ability to
deliver more chemotherapy since neutropenia was not a limiting factor.
In our experience with this regimen at Duke University, virtually
100% of patients develop anemia and more than 80% require transfusions, with an average of
6 to 8 units of blood during six cycles of therapy. With a goal of preventing anemia, we
designed a trial using the same model system in small-cell lung cancer patients receiving
cyclophosphamide, doxorubicin, and etoposide with G-CSF support. Enrolled patients were
previously untreated and were required to have an hematocrit of greater than 32. The
objectives of this trial were to determine if we could prevent anemia in this population
and to evaluate the safety of concurrent erythropoietin and G-CSF.5
Study Design
Subjects were randomized to receive either erythropoietin or placebo
in the first cycle of therapy. The dosing schedule differed from other trials, with 75
µg/dL of epoietin alfa given daily subcutaneously. Although this alteration made direct
comparability with the thrice-weekly schedule somewhat difficult, it constituted a similar
overall weekly total dose.
The endpoint of this trial was the development of anemia (hematocrit
of less than 32) on or after day 1 of cycle 2. Patients on the placebo arm who met the
endpoint of the study received open-label epoietin alfa at the same dose and schedule
outlined above. This allowed a secondary comparison of a preventive strategy in the
epoietin alfa group vs a treatment strategy in the placebo group. In this single
institutional trial, the sample size required only 24 patients to determine a 50%
difference in the primary endpoint.
Results
In determining if neutropenia worsened as a result of adding
epoietin alfa, we found no difference in neutrophil nadir or recovery by the addition of
epoietin alfa to G-CSF. We also found no difference in cycle 1 or throughout the treatment
in cumulative thrombocytopenia or neutropenia in the group receiving epoietin alfa and
filgrastim vs those receiving filgrastim alone.
As an early indicator of red blood cell effect, we measured
reticulocyte count weekly, since reticulocytopenia can occur when patients are treated
with chemotherapy. Our data showed low absolute reticulocyte counts for the first two
weeks in placebo patients. However, the patients on erythropoietin achieved a significant
improvement so that by mid-cycle, they were making reticulocytes at greater numbers.
The main endpoint of the study was to determine how many patients
who remained on the erythropoietin arm avoided anemia and the need for transfusions. There
were a few early failures in cycle 1 in patients with marrow involvement and a few late
failures in patients that perhaps were due to cumulative marrow suppression in cycles 5/6.
However, the majority of patients using erythropoietin maintained a normal hemoglobin
level throughout six cycles of therapy. By contrast, all patients in the placebo arm
developed anemia or needed transfusions in the first three cycles of chemotherapy.
Although the study was small, it was statistically significant in the final analysis (P<.01).
A secondary endpoint was the time to first transfusion. This
appeared to be delayed and reduced in the preemptive erythropoietin group compared with
the placebo group, but the numbers were too small to reach statistical significance. The
patients on the placebo arm who had all crossed over to erythropoietin had an appropriate
reticulocytosis, but they remained more anemic than the preemptive erythropoietin group
through all six treatment cycles. Bone marrow involvement and a baseline hematocrit level
of less than 38 were negative predictive factors in terms of response to this dose and
schedule of erythropoietin.
Study Conclusions
This study raises several interesting points and suggests that
prophylactic use of low-dose, daily erythropoietin can significantly reduce or delay
chemotherapy-related anemia compared with placebo in this model system of fairly intensive
chemotherapy. Patients who were randomized to placebo received transfusions earlier and
remained more anemic throughout the six chemotherapy cycles, despite subsequently being
treated with open-label erythropoietin.
If patients with baseline hematocrit levels greater than 38 receive
preemptive erythropoietin, our results suggest that anemia or the need for transfusions
will likely be avoided. By contrast, patients in this study with marrow involvement who
received erythropoietin with this dosing schedule all failed. To determine the optimal
dosing schedule of erythropoietin to prevent chemotherapy-related anemia requires a larger
study. However, it appears from our study that the earlier one intervenes with
erythropoietin in patients at high risk for the development of chemotherapy-related
anemia, the more likely a benefit will be gained.
US Multicenter Non-Small-Cell Lung Cancer/Erythropoietin Trial
To further address these issues, a multicenter US trial of epoietin
alfa in patients with non-small-cell lung cancer will be initiated in 1998. Investigative
arms will be epoietin alpha vs supportive care alone, beginning in the first cycle of
chemotherapy, for patients with stage IIIb or IV lung cancer. Patients will be stratified
based on hemoglobin levels of either less than or greater than 12 g/dL. Overall, this
trial will study the efficacy of erythropoietin as a treatment strategy for anemic lung
cancer patients at presentation, as well as a preventive strategy for the nonanemic
population. Both groups will be divided into those receiving supportive care alone, with
transfusion, or untreated. This trial will not have a crossover to erythropoietin, but
patients will be managed with transfusion support as indicated.
A further goal is to investigate whether once-a-week dosing is
effective. Most of the usage up to this point has been at 150 U/kg subcutaneously TIW. In
noncardiac surgery, a subcutaneous weekly administration at 40,000 U appears to be
comparable to 10,000 U three times a week. We learned that the thrice-weekly schedule,
while successful, remains a barrier to use of erythropoietin in the community. Thus, all
patients randomized to erythropoietin in this trial will receive epoietin alfa at 40,000 U
subcutaneously over a week.
The primary endpoint for this trial is quality of life. Additional
endpoints include dyspnea, hemoglobin, transfusion requirements, hospitalization,
survival, and time to disease progression. The results of this trial should better define
and quantitate the clinical benefit of erythropoietin in the lung cancer chemotherapy
population.
References
1. Abels RI. Use of recombinant human erythropoietin in the treatment of anemia in
patients who have cancer. Semin Oncol. 1992;19(3 suppl 8):29-35.
2. Henry DH, Abels RI. Recombinant human erythropoietin in the treatment of cancer and
chemotherapy-induced anemia: results of a double-blind and open-label follow-up studies. Semin
Oncol. 1994;21(2 suppl 3):21-28.
3. Glaspy J, Bukowski R, Steinberg D, et al. Impact of therapy with epoietin alpha on
clinical outcomes in patients with nonmyeloid malignancies during cancer chemotherapy in
community oncology practice. J Clin Oncol. 1997;15:1218-1234.
4. Crawford J, Ozer H, Stoller R, et al. Reduction by granulocyte colony-stimulating
factor of fever and neutropenia induced by chemotherapy in patients with small cell lung
cancer. N Engl J Med. 1991;315:164-170.
5. Crawford J, Blackwell S, Shoemaker D, et al. Prevention of chemotherapy-related
anemia by recombinant human erythropoietin (EPO) in patients with small cell lung cancer
receiving cyclophosphamide, doxorubicin, and etoposide (CAE) chemotherapy with G-CSF
support. Lung Cancer. 1997;18(suppl 1):205.
DR BENNETT
I wonder whether a better trial design for the proposed multicenter
study would be to offer EPO to both arms, with one group receiving the standard regimen
(150 U/kg three times a week) and the other receiving 40,000 U/wk, with a crossover
potential. I think it will be difficult to offer a population of patients the opportunity
to participate in a trial in which they may not receive a benefit.
DR CRAWFORD
I agree, but we have the largest database of patients who have
received EPO -- 2,000 patients -- that I know of for any cancer other than breast cancer
adjuvant therapy. There are another 2,000 patients from another trial. There were good
cross-sectional data from the registration trial, but if you look at clinical practice, a
minority of cancer patients currently receive EPO when they develop anemia.
DR BENNETT
I think that is changing. The data at our center indicate that
increasing numbers of patients are being given EPO. Except in the radiation therapy arena,
where they are getting it in a different setting, most patients are getting it after the
recognition of severe anemia, but some clinicians are beginning to use EPO to prevent
anemia.
DR CRAWFORD
Our new study provides a way to define the population up front, to
show the magnitude of difference, and to look at an endpoint other than transfusion
requirements. To date, the primary clinical endpoint of most EPO trials has been to reduce
transfusion requirements. Making quality of life a primary endpoint instead may provide
more important data for clinical efficacy. Also, the trial design, as it is, is simple,
and crossovers -- while they sometimes answer additional questions -- always complicate
evaluations.
DR SABA
An important aspect of this presentation has been the quality of
life data. If there is a definite difference between the transfused patient vs the EPO
patient in terms of improvement of fatigue or general well being, then we have to look at
other patient populations, too.
DR CRAWFORD
My own bias is that there may be some other effect of
erythropoietin. The main issue is that you maintain that hemoglobin of 10 for only a short
time with transfusion. Dr Itri has shown a nice linear increase with each gram of
hemoglobin, from 8 to 10, 9 to 10, to 11 to 12. And there is just as much, if not more,
improvement in that 10-to-12 range as there is from 8 to 10. I think that is something we
have not appreciated.
DR ZUCKERMAN
I have tried to make a similar point in the past -- that it costs
some extra units of blood to get patients' hemoglobins up several points, but once they
are up to your chosen level, it costs no more red cells to keep them between 10 and 12 g
of hemoglobin than it does to keep them between 7 and 9 or 6 and 8 g of hemoglobin. And
people will do better with that initial investment. Erythropoietin is another (and
probably better) way to accomplish that goal. I think this kind of a study is important to
answer the question of whether transfusions or EPO should be given to a patient with a
hemoglobin of 10 g/dL.
DR MOSCINSKI
We need to separate red cell transfusion from erythropoietin. Much
of the argument in the community, as well as a lot of the attitude, has to do with the
fact that while red cell transfusions are safe, they have several other complications and
comorbid features. Erythropoietin may not necessarily have these.
DR SABA
Another issue is that it takes about two to three weeks for the
erythropoietin to take effect. So instead of starting the erythropoietin simultaneously
with the chemotherapy, you may start the erythropoietin earlier. In our hematologic
malignancy patient population, we realized that if we had started erythropoietin earlier
than the chemotherapy, patients could have derived more benefit because all the bone
marrow was packed with erythroid precursors.
DR CRAWFORD
I think the idea of the community study was to try to emulate
practice. We will administer erythropoietin once a week rather than three times a week to
make it easier for practices. I do not know how asking them to start a week before the
chemotherapy would translate into practice. It may be something we can look at later, but
I think we will start with the first treatment.
DR ZUCKERMAN
I would like to amplify the iron supplementation issue. During the
course of our studies in patients with renal failure, we discovered that erythropoietin is
an effective way to draw iron out of the iron stores. In fact, another use for
erythropoietin was being discussed and considered -- to help to mobilize the iron in
hemochromatosis patients.
DR SABA
There are some reports of effective use of EPO in mobilizing iron
from patients with hemochromatosis. In our clinic, we have four or five patients with
hemochromatosis whom we have treated more effectively with EPO than with anything else.
DR SPIVAK
At this moment, in dialysis units and elsewhere, the issue of how to
deal with iron in patients with renal failure who are getting erythropoietin is being
analyzed. Dialysis patients lose blood at a tremendous rate -- they are essentially
chronic bleeders -- and erythropoietin therapy can really put stress on the system. I
would suggest that it is an entirely different situation you are facing in patients who do
not have renal
disease. Patients with cancer, rheumatoid arthritis, HIV infection, or a variety of other
illnesses probably do not need iron. I think it was Mark Goldberg's group in Boston that
showed if the serum ferritin in patients was greater than 100, you reached maximal
efficacy with erythropoietin, as measured by reticulocyte rates or blood production.
Ferritin levels higher than that did not provide greater effect when giving
erythropoietin.
DR CRAWFORD
There is a proviso in our new study design. Patients with ferritin
levels of less than 100 at baseline go on iron from the start. However, for those who do
not achieve improvements in their hemoglobin or for the groups that fail, the question of
mobilizing enough iron arises. Hopefully, you will not shift so much iron into the red
cell mass that they become functionally iron deficient.
DR SPIVAK
You will mobilize iron. This has been shown in patients with
rheumatoid arthritis who have the classical anemia of chronic disease with a low serum
iron, a low iron-binding capacity, a low percentage of saturation, and a high serum
ferritin level. They respond to erythropoietin unless other factors affect the response,
such as inflammation so intense that they cannot. However, it is not the iron. If they
have adequate iron stores, they will respond.
DR LEE
I formerly used this logic to avoid iron
supplementation in our chemoradiation therapy patient population. In our first four
patients treated with erythropoietin but without iron supplementation, hemoglobin levels
dropped during chemoradiation therapy. We then changed the protocol to include oral
ferrous sulfate for 12 patients. All but one maintained hemoglobin at more than 10 g%, and
their median hemoglobin nadir was 11.8 g%. There was a tremendous effect of iron
supplementation in our patient population. In lung cancer, we do not expect iron
deficiencies, not like the gastrointestinal or gynecologic cancers.
From the Thoracic Oncology Program, Duke University Medical Center, Durham, NC.
Address reprint requests to Jeffrey Crawford, MD, Duke University
Medical Center, Director of Clinical Research, Division of Medical Oncology and
Hematology, Duke Comprehensive Cancer Center, Durham, NC 27710.
Dr Crawford is a speaker/consultant and receives research support
from Ortho Biotech, Inc.
Back to Cancer Control Journal Supplement Volume 5 Number 2