DOI: 10.1111/sdi.

12841

REVIEW ARTICLE

Anemia management in cancer patients with chronic kidney

disease

Sheron Latcha

Renal Division, Department of Medicine,

Memorial Sloan Kettering Cancer Center, Abstract
New York, New York Cancer and kidney disease are linked by causality and comorbidities. Observational
Correspondence data show an increased risk of malignancy as renal function declines. Erythropoietin
Sheron Latcha, MD, FASN, Associate stimulating agents (ESAs), which are the cornerstone therapy for anemia patients
Attending, Renal Division, Department of
Medicine, Memorial Sloan Kettering Cancer with chronic kidney disease and cancer, are associated with increased risks for can‐
Center, 1275 York Avenue, New York, NY cer, cancer‐related mortality, progression of disease, and thromboembolic events.
10065.
Email: [email protected] This article examines the recently published guidelines for ESA use in cancer patients
from the American Society of Clinical Oncology and American Society of Hematology
and attempts to contextualize them to the care of patients with coexistent CKD, can‐
cer, and anemia.

Cancer and kidney disease are linked by causality and comorbidi‐ cancer, 2 and in patients with cancer, ESA use correlated with tumor
ties. More than 50% of patients with a diagnosis of cancer or ESRD growth, shorter progression‐free survival (PFS),3 and an increased
also have anemia. Patients with cancer can develop CKD due to di‐ risk of mortality.4-6
rect and indirect effects of the tumor on the kidney itself or from Observational data show an increased risk of malignancy in ESRD
treatment‐related toxicity. Erythropoietin‐stimulating agents (ESAs) patients.7 Secondary analysis of the Trial to Reduce Cardiovascular
form the cornerstone for treatment of anemia caused by CKD or Events with Aranesp Therapy (TREAT) 8 demonstrated that patients
cancer therapy. Since they were first introduced in 1989, ESAs have with a previous malignancy who received darbepoetin to target a
significantly decreased transfusion dependence and the attendant Hg of 13 g/dL had a significantly higher risk of dying from cancer
risks for infection, iron overload, and presensitization and they have than those in the placebo group. Because patients with cancer are
improved anemia‐related symptoms and quality of life measure in excluded from CKD trials, and patients with CKD are excluded from
patients with either CKD or cancer. cancer trials, there are no well‐designed clinical studies that examine
Over a decade after their introduction, data linking ESA use with the relationship between ESA exposure in patients with CKD and the
negative clinical outcomes began to emerge. Prior to 1998, placebo‐ risk for new or recurrent malignancy. The Surveillance of Epoetin‐
controlled studies of CKD patients were designed to evaluate the Adverse Events of Stroke and Cancer (SEASCAN)9 study, designed
short‐term safety and efficacy of erythropoietin (EPO); most tar‐ to examine the risk for cancer with ESA use, found no significant
geted a Hg of 9.5‐12 g/dL. Although not designed to evaluate car‐ association between ESA treatment and malignancy. Regrettably,
diovascular endpoints, these studies demonstrated an increased follow‐up was limited to 6 months, making it difficult to draw any
risk for hypertension and thrombosis in the ESA‐treated groups. substantive conclusions.
Between 1998 and 2008 a shift occurred during which larger stud‐ Currently, there are no clear and specific guidelines for ESA
ies, principally sponsored by ESA manufacturers, compared higher use in the management of anemia in patients with both CKD and
versus lower Hg targets (9‐12 g/dL vs 12‐15 g/dL). A meta‐analy‐ a current or past diagnosis of cancer. In 2012, the Kidney Disease:
ses of ESA trials in CKD patients revealed that higher Hg targets Improving Global Outcomes (KDIGO)10 gave a grade IB recommen‐
were associated with increased risk for stroke, hypertension, and dation (“we recommend” with “moderate” quality of evidence”) that
vascular access thrombosis across all stages of CKD.1 ESA use also ESAs be used with “great caution” in all CKD patients with active
corresponded with a higher odds ratio (OR) for developing a new malignancy, particularly when cure is the anticipated outcome. For

Seminars in Dialysis. 2019;32:513–519. wileyonlinelibrary.com/journal/sdi

© 2019 Wiley Periodicals, Inc. | 513
514 | LATCHA

CKD patients with a past history of malignancy, KDIGO made a sim‐ and activator of transcription (STAT) 5, protein kinase B (AKT) and
ilar but weaker recommendation (grade 2C: “we suggest” with “low” extracellular signal‐regulated kinase (ERK) 1/2. Stimulation of these
quality of evidence). In June 2019, the American Society of Clinical proteins can lead to cellular differentiation, proliferation, and anti‐
Oncology (ASCO) and the American College of Hematology (ASH) apoptosis.12 EPO binding to the EpoR also induces nuclear translo‐
updated their clinical practice guidelines for the management of can‐ cation of nuclear factor‐κB (NF‐ κB), which may induce anti apoptotic
11
cer‐associated anemia (Table 1). The objective of this article was to gene transcription.13
review some of the clinical data upon which these guidelines were Multiple tumor types express the EpoR, including lung, breast,
established. I hope that this background information will enable cli‐ colon, gastric, squamous cell of the head and neck, and uterine can‐
nicians to make more informed decisions in the application of these cer.14-19 A possible role for JAK‐STAT signaling in cancer progression
guidelines to their specific patients. was demonstrated in studies on cell lines from squamous cell car‐
In vitro and in vivo data indicate that ESAs promote tumori‐ cinoma of the head and neck. EPO promoted tumor invasion was
genesis and angiogenesis. Like endogenous EPO, ESAs bind to and subsequently blocked by a JAK inhibitor as well as in a cell line with
activate extracellular EPO receptors (EpoR) on erythroid progeni‐ a STAT5A mutation.18 Tumor migration in a breast cancer cell line
tor cells in the bone marrow. When ESAs bind the EpoR on the was increased through the activation of EPO/EpoR induced ERK ac‐
cell membrane, there is autophosphorylation of Janus‐activated 2 tivation. This effect was mitigated in the presence of soluble EpoR
(JAK2) kinase, a transducer of cancer cell signaling. In turn, JAK2 or anti‐Epo antibodies. 20 In human melanoma cells, EPO activated
kinases phosphorylate tyrosine residues on the intracellular do‐ the AKT signaling pathway and increased tumor cell survival. Tumor
mains of the EpoR. The intracellular domains act as docking sites cell viability was diminished following cotreatment with an AKT
for various cytoplasmic signaling proteins such as signal transducer inhibitor. 21
While these data suggest a protumorigenic role for ESAs, these
TA B L E 1 Summary of the American Society of Clinical pleiotropic molecules do not always behave as anticipated and can
Oncology/American Society of Hematology Clinical Practice
be difficult to properly study with currently available methods. A
Guideline Update
major limitation of current studies is that the commercially available
Before offering an ESA, conduct a history, physical exam and antibodies used to target the EpoR are nonspecific and overlap with
diagnostic tests to identify alternative causes of anemia aside from
other proteins. Heat shock protein (HSP)70 is one such overlapping
chemotherapy or an underlying hematopoietic malignancy.
protein and high expression of HSP‐70 is itself linked to poorer prog‐
ESAs should not be offered to patients with chemotherapy associ‐
nosis, more aggressive disease and resistance to chemotherapy. 22
ated anemia whose cancer treatment is curative in intent.
Importantly, there may be considerable reporting biases since most
ESAs may be offered to patients with chemotherapy associated
of the negative studies for evidence of functional EpoR expression
anemia whose cancer treatment in not curative in intent and whose
Hg is <10 g/dL. Depending on the severity of anemia and clinical in tumor cells were funded by ESA manufacturers, whereas positive
circumstances, RBC transfusion is also an option. studies come from nonfunded academic researchers. 23-26
ESAs should not be offered to patients with nonchemotherapy Based on the results of the in vitro and in vivo studies which
associated anemia. One exception is that ESAs may be offered to demonstrated that the EPO‐EpoR complex promotes differentiation
patients with lower risk MDS and a serum EPO level <500 IU/L
of normal endothelial cells into vascular tubes and new blood ves‐
For patients with MM, NHL or CLL, clinicians should observe the sel formation,17 researchers hypothesized that ESAs’ proangiogenic
hematologic response before considering an ESA.
effects and their capacity to increase Hg would increase delivery
All ESA (epoetin beta and alfa, biosimilar epoetin alfa) are consid‐
of chemotherapy, decrease the hypoxic tumoral milieu, and thereby
ered equivalent with regard to safety and efficacy.
improve patient outcomes. Consequently, several clinical cancer tri‐
ESAs increase the risk for thromboembolism. Physicians must
als were designed to target Hg levels above 12 g/dL. Contrary to
weight the risks of thromboembolism and use caution and clinical
judgment when considering ESA use. the expected outcomes, ESA use in several of these trials demon‐

When starting or modifying ESA doses, follow the FDA guidelines. strated greater locoregional progression, shorter PFS, lower odds of
survival, shorter disease‐free survival and increased death.3,5,27-29
ESAs may be used to target the lowest Hg concentration needed to
avoid or reduce the need for RBC transfusions. These discrepant findings may be due to the inability to adequately
ESAs should be discontinued in patients who do not respond within recreate the tumor microenvironment or because tumor vasculature
6‐8 wk, as evidenced by a rise in Hg of less than 1‐2 g/dL or no is aberrant and is regulated by different physiologic signals.
decrease in transfusion requirement. Patients who do not respond A Cochrane Database analysis published in 201230 on ESA
to ESA should be reevaluated for underlying tumor progression,
treatment in cancer patients showed that ESAs significantly in‐
iron deficiency or other etiologies.
creased mortality (HR 1.17) and worsened overall survival (HR
Iron replacement may be used to improve Hg response and reduce
1.06). Importantly, an analysis restricted to trials of patients receiv‐
RBC transfusions for patients receiving ESA with or without iron
deficiency ing chemotherapy found only an insignificant trend toward higher
overall mortality (OR 1.04). It was only when the analysis was lim‐
Abbreviations: CLL, Chronic Lymphocytic Leukemia; EPO, Epoetin; ESA,
Erythropoietin Stimulating Agents; MDS, Myelodysplastic Syndrome; ited to trials with patients not on cancer therapy that ESA use sig‐
MM, Multiple Myeloma; NHL, Non Hodgkin Lymphoma. nificantly increased mortality (OR 1.23). Additional meta‐analyses
LATCHA | 515

TA B L E 2 Guidelines for epoetin and darbepoetin in adults can be accessed at the webpages for the respective societies.10,11
Both guidelines propose that an initial investigation for anemia
Erythropoietin alfa Darbepoetin alfa
should include a history, physical exam, and diagnostic tests (CBC,
Starting dose
reticulocyte count, serum iron, ferritin and iron saturation [TSAT],
40 000U SC weekly 500 mcg Q 3 weeks B12, and folate levels), to identify causes of anemia other than che‐
Reduce motherapy, a hematopoietic malignancy or CKD. The ASCO/ASH
By 25% By 40% guidelines (Table 1) refer largely to solid tumors.
Hg increases > 1 g/dL in any 2 week period or to a level at which a A review of these guidelines highlights the need for an ongoing
transfusion can be avoided dialog between the nephrologist and oncologist since decision mak‐
Increase ing depends on patient specific factors. Implicit in the guidelines is
To 60 000U SC weekly if Hg in‐ N/A the need for the physician to discuss the relative risks and benefits
creases <1 g/dL after 4 weeks and of ESAs versus RBC transfusion (tRBC) at treatment initiation and
remains <10 g/dL
when changes in therapy occur. The ASCO/ASH guidelines specify
Withhold
that ESAs should be used only when the anemia in a cancer patient
Hg exceeds a level needed to avoid Hg exceeds a level is the result of myelosuppressive therapy, that is, any cancer treat‐
transfusion. Restart at 25% below needed to avoid
ment, including radiation, which kills normal cells and cancer cells
the previous dose when the Hg ap‐ transfusion. Restart
proaches a level where transfusions at 40% below the in the bone marrow. ESAs should only be used if the Hg is <10 g/
may be required. previous dose when dL and ESAs should not be used in cancer patients who are not on
the Hg approaches active treatment and for whom cancer treatment is expected to cure
a level where
the disease.
transfusions may be
required. For patients with symptomatic anemia, KDIGO and ASCO/ASH

Discontinue agree that tRBC will immediately improve symptoms, whereas ESAs
can take weeks to months to provide relief. For patients with asymp‐
After 8 weeks if no response as measured by Hg level or If
transfusions are still required. tomatic anemia, when deciding whether to treat with tRBCs or ESAs,
the nephrologist must take into account the patient's primary cancer
diagnosis, life expectancy from the cancer, risk of TE, transfusion
of controlled trials comparing patients on cancer therapy with and history, comorbidities, quality of life, and risks attendant to tRBCs.
without ESAs have failed to show significant differences in disease The Cochrane Database analysis showed that ESA use does, as as‐
progression, tumor response rates, or odds of survival between the sumed, significantly decrease the risk for PRBC transfusion.30 Of
two groups.30-35 Several additional meta‐analyses, found insufficient note, 1 unit of PRBC was considered significant in this analysis. One
evidence that ESAs have any effect on progression of disease.33,36,37 unit of RBCs may not carry any meaningful risks for viral infection or
Akin to findings in CKD patients, ESA use is associated with in‐ iron overload in a cancer patient with CKD who has a limited life ex‐
creased risk for thromboembolic events (TE) in cancer patients. A pectancy because of one or both diagnoses. Among ESRD patients
meta‐analyses in 2013 showed that the rate of TE was higher in age >65 years, 30%‐54.5% of patients will die within 1‐year after
ESA‐treated patients (RR 1.51). There were fewer TE when ESA dialysis initiation. This rate is as high as 73% for patients who are de‐
treatment was delayed until the baseline Hb was less than 10 g/dL.31 pendent on assistance for their activities of daily living.41,42 Having
There is no information on what Hg targets are associated with no an active malignancy or metastatic disease is associated with con‐
added mortality risk. sistently worse 6‐month and 1‐year survival in ESRD patients.43,44
In composite, the results of trials on ESA use in cancer and CKD With the current level of donor screening, tRBC is associated
populations suggest that ESA use is associated with increased mor‐ with a miniscule risk of viral infections, currently 1:1 million, 1:1.2
tality and TE and that this increased risk was associated with higher million and 1:1.5 million for Hepatitis B, Hepatitis C and HIV, re‐
Hg targets. The mortality risk with ESAs is significantly greater spectively.45,46 Iron overload from tRBC is also unlikely to be a con‐
when patients are not receiving cancer therapy. Whether adverse sideration in all but a few cancer patients with ESRD. Based on the
outcomes are related to the total dose of ESAs or to the targeted approximation that each unit of RBCs has about 250 mg of iron, it's
or achieved Hg has been examined in CKD. In a meta‐analysis38 generally accepted that transfusion of 15‐20 units of PRBC will re‐
39 40
and post hoc analyses of the CHOIR and TREAT trials, higher sult in signs of iron overload. It remains unknown if CKD patients
ESAs doses were associated with increased risk for mortality and with functional iron deficiency will manifest symptoms of iron over‐
cardiovascular endpoints, independent of target or achieved Hg. load with fewer units of tRBCs. Nonetheless, for patients with a
However, these results may have been influenced by an indication relatively short life expectancy, single or repeated tRBC generally
bias toward higher ESA doses among patients with ESA hypore‐ will not engender clinically significant sequelae. For those who have
sponsiveness due to greater comorbidities and inflammation. bone marrow failure, are ESA unresponsive, and who become iron
The complete guidelines for anemia management in CKD pa‐ overloaded is due to transfusion dependence, chelation therapy may
tients from KDIGO and in cancer patients from the ASCO and ASH be required.
516 | LATCHA

Practical considerations will also influence the decision regard‐ cardiovascular disease and additional comorbidities, it may be
ing the use of ESAs or tRBC. Since most outpatient dialysis units will reasonable to consider initiation of ESAs between 8 and 8.5 mg/
not perform tRBC, this prescription requires a visit to another treat‐ dL or at a threshold value where the patient is known to become
ment facility as well as additional venipuncture, both of which can be symptomatic.
difficult in ESRD patients. The presence of volume overload, history As previously stated, meta‐analyses demonstrate an approx‐
of transfusion reactions, and religious objections to blood products imately 50% increase risk of TE in patients with cancer receiving
would argue against tRBC. Insurance coverage for ESA therapy may ESAs with fewer TE when ESA treatment was delayed until the
be an issue for some patients. baseline Hb was <10 g/dL. The absolute pooled event rate in the
Despite data indicating that ESA use in patients on chemother‐ treatment and control arms were 5.8% and 3.2%, respectively, with
apy does not increase mortality risk or progression of disease, and ranges of 0%‐30.8% and 0%‐14.5%, respectively.31 Notably, specific
the absence of any study or meta‐analyses of ESA treatment out‐ risk factors for TE were not identified in these analyses and there
comes by subgroups based on treatment intent, ASCO/ASH recom‐ was no threshold level below which no risk was evident. Based on
mends that ESA not be offered to patients whose cancer treatment these findings, ASCO/ASH guidelines nebulously recommend ex‐
is intended to cure. KDIGO recommends “caution” when cure is ercising caution when ESAs are used concomitantly with treatment
anticipated. Advances in precision medicine and biologic therapies strategies and diseases where the risk of TE is increased and that the
have bolstered disease‐free intervals and progression free survival, clinician must weigh the risks of TE and benefits of ESA in an individ‐
making clinical outcomes from cancers a moving target such that ual patient. Additionally, ESA therapy can be considered when the
the distinction between treatment aimed at “palliation” or “cure” Hg is <10 g/dL but should probably not be used until substantially
is not always categorical. For example, the FDA recently granted below this level and then only at lowest dose to minimize symptom‐
approval for two CAR T‐cell therapies 47,48 based on clinical trials atic anemia and tRBC.
that demonstrated overall response and complete remission rates Epidemiological studies show that tumors of the pancreas,
of 80% and 60%, respectively, in patients who previously had re‐ brain, lung, and ovary are associated with the highest risk of TE.41,55
sistant and multiply relapsed hematologic malignancies prior to the Relatively low risks are observed with breast and prostate cancer.55
biologic therapy. Surgery and hormonal therapy are both associated with increased
For a patient whose cancer is potentially curable and who may risk of TE in cancer patients. Of note these treatments are often used
become a candidate for a renal transplant in the future, the hazards for the management of breast and prostate cancer. Antiangiogenic
associated with allosensitization would seem to favor ESAs. Two drugs are also associated with an increased risk for TE.55 In gen‐
database analyses of matched cohorts of patients awaiting primary eral, cancer types that are biologically aggressive as manifested by
renal transplant revealed that, when compared to nontransfused short survival time and early metastatic spread are correlated with
patients, transfused patients have clinically significant increases in a higher incidence of thrombosis.56 Across a range of cancer types,
49,50
HLA antibody levels and panel reactive antibody. Currently, it metastatic disease at the time of cancer diagnosis was found to be
remains unclear if leukoreduction reduces the incidence of allosen‐ the strongest predictor of subsequent venous thrombosis.57 In gen‐
51,52
sitization from tRBC. Thus the choice of tRBC over ESA may eral, the incidence of TE is highest within the first 3 months of can‐
reduce the likelihood of transplantation. In general, 2 years is the cer diagnosis and remains elevated but relatively decreased between
minimum disease‐free time interval after treatment of a cancer for a 3‐12 months. The lower risk observed beyond 1 year may be due
patient to become eligible for a renal transplant. Exceptions to this to response to cancer treatment or to patients succumbing to their
time criteria include breast cancer beyond in situ lesions, malignant disease.58
melanoma, node positive colorectal cancer, and invasive cervical Based on these data, it may be prudent to avoid ESAs in patients
cancer, for which some societies recommend a 5‐year interval. 53,54 with a newly diagnosed malignancy, a history of TE, metastatic dis‐
Patients who are healthy enough to be listed for a renal transplant ease at presentation, tumors with aggressive features, or with any
after being cured of their cancer may still have a significant wait on treatment that includes surgery, hormonal or antiangiogenic ther‐
a transplant list if they do not have a live donor. Thus, for cancer apy. Overall, the risk for TE is greater among inpatients, but the ma‐
patients who have a reasonable expectation to become a viable jority of TE occur in outpatients (about 80%) because most patients
renal transplant candidate in the future, minimizing the risk of both are treated in outpatient settings.59 The ASCO/ASH guidelines on
allosensitization associated with tRBC as well as any potential protu‐ ESA use do not include a discussion on thromboprophylaxis. Those
morigenic role for ESAs are appropriate. recommendations were made in a separate document60 in which
ASCO/ASH do not provide a nadir Hg value at which ESAs can ASCO/ASH recommends initiating pharmacologic thromboprophy‐
be started. For ESRD patients, KDIGO recommends avoiding Hg laxis with oral anticoagulants or low molecular weight heparin prior
levels below 9g/dL, presumably because cardiovascular disease to starting systemic chemotherapy in patients at high risk for TE,
is highly prevalent among ESRD patients. It's been my experience unless there is a clinical contraindication. To determine risk, ASCO
that younger patients and those without significant cardiovascular recommends using the Khorana61 score and to begin thrombopro‐
disease or numerous comorbidities can reasonably tolerate Hg val‐ phylaxis for patients on systemic chemotherapy who have a score
ues as low as 7‐7.5 g/dL. For older patients or those with significant of 2 or higher. A Hg below 10g/dL and ESA use each alone confer
LATCHA | 517

a score of 1. Tumors of the pancreas and stomach have a score of 2 Increased HIF expression in a range of human cancers is asso‐
and gynecologic, genitourinary (except prostate), and lung cancers ciated with poorer prognosis and outcomes.69 It remians unclear if
have score of 1. HIF is acting as a tumor promoter or if higher HIF levels just reflect a
ASCO/ASH makes one exception to its recommendation that more hypoxic milieu in faster growing and more aggressive tumors.
ESAs be used only for patients who are on palliative cancer therapy. While there is a theoretical concern that HIF‐PHIs may be protumor‐
It has been demonstrated that patients with myelodysplastic syn‐ igenic, there are no clinical data that demonstrate causality. In fact,
drome (MDS) and a serum EPO level ≤500 IU/L had marked increases everolimus and temsirolimus, which are both indirect HIF inhibitors
in Hg levels with ESAs.62 Nephrologists do not routinely check serum in the class of mTOR inhibitors, are effective against several tumor
EPO levels in CKD patients but ASCO/ASH suggests that EPO levels types, including renal cell carcinoma. There are several ongoing
be checked in patients with lower risk MDS. For patients with other Phase II and III clinical trials of HIF inhibitors across several tumor
hematologic malignancies like myeloma, non‐Hodgkin lymphoma, or types65 which may help to better clarify the complicated interplay
chronic lymphocytic leukemia who are receiving concurrent myelo‐ between HIF inhibitors and tumorigenesis.
suppressive therapy, it is recommended that the clinician evaluate In summary, the guidelines for anemia management in patients
the hematologic response to cancer treatment before considering with cancer and CKD are not based on high‐quality experimental
ESA use. or clinical data that verifiably demonstrate a causal relationship
If the decision is made to begin ESAs, it is recommended that the between ESAs, CKD, cancer, and TE. ASCO/ASH states that ESAs
clinician follow the FDA dosing guidelines (Table 2). With respect to should only be used to treat anemia that is the result of myelosup‐
efficacy and safety, EPO beta and alpha, darbepoetin, and biosimilar pressive therapy in cancer patients. This narrowly defined indication
epoetin alfa are considered equivalent. A meta‐analysis showed that for ESA use in cancer patients is not meaningful for clinicians car‐
up to 46% of patients with no rise in Hg by 2‐4 weeks will ultimately ing for patients with CKD and cancer who are on myelosuppressive
respond to ESAs.63 However, if a patient's Hg does not increase therapy since it is impossible to determine the exact etiology of ane‐
1‐2 g/dL, or their need for blood transfusions is not decreased after mia in such patients. The diagnostic value of serum EPO levels in
6‐8 weeks on ESAs, this is considered an ESA treatment failure and patients with advanced CKD is of limited benefit.70 Moreover, the
the ESA should be discontinued. At that time, such patients should results of several meta‐analyses suggest no increased risk for can‐
be reevaluated for causes of ESA resistance like tumor progression, cer‐related complications if patients are receiving ESAs on cancer
blood loss, marrow replacement by disease or marrow suppression therapy.
from medications, and infection. For patients with symptomatic anemia, tRBCs seems to be the
Clearly there are numerous gaps in our knowledge of anemia most prudent intervention. For patients with asymptomatic anemia.
management in patients with CKD and a prior or present cancer a more tactical approach may provide greater benefit and minimize
diagnosis. Hypoxia‐Inducible Factor Prolyl Hydroxylase Inhibitors harm. It may be best to avoid tRBC for patients who may require a
(HIF‐PHI) are a novel class of orally available molecules and rep‐ renal transplant in the future if you think that the patient will survive
resent a new therapeutic strategy for anemia in CKD. This in‐ with no evidence of disease for long enough to qualify for a trans‐
novation will likely import even greater uncertainty for anemia plant. For patients at high risk for TE (recent surgery, immobility, or
management for CKD patients with cancer. Currently, there are who are receiving anti angiogenic or hormonal therapy), it may be
5 HIF‐PHI products being investigated in human subjects with reasonable to begin thromboprophylaxis with an oral anticoagulant
nondialysis‐dependent and dialysis‐dependent CKD. Targeting the or low molecular weight heparin prior to starting cancer therapy and
HIF pathway with HIF‐PHIs may help to circumvent functional iron then consider ESA use after there has been some response to treat‐
deficiency as a result of elevated hepcidin levels. In clinical trials, ment. For patients with a poor functional status, significant comor‐
HIF‐PHIs induce erythropoiesis in the presence of normal oxygen bidities, and whose cancer prognosis is poor, tRBCs and ESA may
tension, decrease hepcidin levels and consequently increase the both be reasonable approaches to anemia management depending
bioavailability of iron. 64 on patient preference and comfort. ESAs should probably be started
The HIF protein was found 1991 next to the EPO gene during at a Hg level at which the patient has not developed symptoms re‐
the early studies on ESAs. In response to hypoxia, HIF‐1α expres‐ lated to anemia but below which the clinician anticipates that the
sion is up‐regulated and the molecule translocates from the cyto‐ patient will become symptomatic. For younger patients without
plasm to the cell nucleus where it forms a functional heterodimer significant comorbidities, this may be as low as 7.5 g/dL. For older
with the HIF‐β subunit. This complex acts as a transcriptional factor patients and those with cardiovascular disease, 8‐8.5 g/dL may be a
for hundreds of genes involved in angiogenesis, including vascular reasonable threshold.
endothelial growth factor (VEGF), platelet‐derived growth factor
(PDGF), and angiopoietin‐1 (ANGPT1).65 In addition to regulating
REFERENCES
the expression of genes involved with tumor growth, invasion, and
metastasis,66 the HIF pathway plays a key role in the development of 1. Palmer SC, Navaneethan SD, Craig JC, et al. Meta‐analysis: erythro‐
resistance to anti‐cancer therapies67 and HIF‐1α facilitates escape of poiesis‐stimulating agents in patients with chronic kidney disease.
Ann Intern Med. 2010;153(1):23‐33.
tumor cells from T‐cell recognition.68
518 | LATCHA

2. Rene E, Lazrak HH, Laurin LP, Elftouh N, Vallee M, Lafrance JP. 20. Lester RD, Jo M, Campana WM, Gonias SL. Erythropoietin pro‐
Association of erythropoiesis‐stimulating agents and the in‐ motes MCF‐7 breast cancer cell migration by an ERK/mitogen‐
cidence risk of cancer diagnosis among chronic dialysis pa‐ activated protein kinase‐dependent pathway and is primarily
tients: a nested case‐control study. Nephrol Dial Transplant. responsible for the increase in migration observed in hypoxia. J Biol
2017;32(6):1047‐1052. Chem. 2005;280(47):39273‐39277.
3. Henke M, Laszig R, Rübe C, et al. Erythropoietin to treat head 21. Kumar SM, Yu H, Fong D, Acs G, Xu X. Erythropoietin activates the
and neck cancer patients with anaemia undergoing radiother‐ phosphoinositide 3‐kinase/Akt pathway in human melanoma cells.
apy: randomised, double‐blind, placebo‐controlled trial. Lancet. Melanoma Res. 2006;16(4):275‐283.
2003;362(9392):1255‐1260. 22. Sizer KC. Heat shock protein 70, erythropoietin, and cancer. J Clin
4. Wright JR, Ung YC, Julian JA, et al. Randomized, double‐blind, pla‐ Oncol. 2007;25(27):4326; author reply 4326‐4327.
cebo‐controlled trial of erythropoietin in non‐small‐cell lung cancer 23. Hardee ME, Kirkpatrick JP, Shan S, et al. Human recombinant eryth‐
with disease‐related anemia. J Clin Oncol. 2007;25(9):1027‐1032. ropoietin (rEpo) has no effect on tumour growth or angiogenesis. Br
5. Leyland‐Jones B, Semiglazov V, Pawlicki M, et al. Maintaining J Cancer. 2005;93(12):1350‐1355.
normal hemoglobin levels with epoetin alfa in mainly nonanemic 24. Swift S, Ellison AR, Kassner P, et al. Absence of functional EpoR ex‐
patients with metastatic breast cancer receiving first‐line chemo‐ pression in human tumor cell lines. Blood. 2010;115(21):4254‐4263.
therapy: a survival study. J Clin Oncol. 2005;23(25):5960‐5972. 25. Bennett CL, Lai SY, Sartor O, et al. Consensus on the existence of
6. Bohlius J, Schmidlin K, Brillant C, et al. Recombinant human functional erythropoietin receptors on cancer cells. JAMA Oncol.
erythropoiesis‐stimulating agents and mortality in patients 2016;2(1):134‐136.
with cancer: a meta‐analysis of randomised trials. Lancet. 26. Bennett CL, Lai SY, Henke M, Barnato SE, Armitage JO, Sartor O.
2009;373(9674):1532‐1542. Association between pharmaceutical support and basic science
7. USRDS. United States Renal Data System: chapter 5 mortality. research on erythropoiesis‐stimulating agents. Arch Intern Med.
2017; https​://www.usrds.org/2017/downl​oad/v2_c05_Morta​ 2010;170(16):1490‐1498.
lity_17.pdf. Accessed March 1, 2019. 27. Hedenus M, Adriansson M, Miguel JS, et al. Efficacy and safety of
8. Pfeffer MA, Burdmann EA, Chen C‐Y, et al. Baseline characteristics darbepoetin alfa in anaemic patients with lymphoproliferative ma‐
in the Trial to Reduce Cardiovascular Events With Aranesp Therapy lignancies: a randomized, double‐blind, placebo‐controlled study.
(TREAT). Am J Kidney Dis. 2009;54(1):59‐69. Br J Haematol. 2003;122(3):394‐403.
9. Imai E, Yamamoto R, Suzuki H, Watanabe T. Incidence of symptom‐ 28. Untch M, Fasching PA, Konecny GE, et al. PREPARE trial: a random‐
atic stroke and cancer in chronic kidney disease patients treated ized phase III trial comparing preoperative, dose‐dense, dose‐inten‐
with epoetins. Clin Exp Nephrol. 2010;14(5):445‐452. sified chemotherapy with epirubicin, paclitaxel and CMF versus a
10. Kidney Disease: Improving Global Outcomes Chronic Kidney standard‐dosed epirubicin/cyclophosphamide followed by pacli‐
Disease Guideline Development Work Group M. KDIGO Clinical taxel +/‐ darbepoetin alfa in primary breast cancer–results at the
Practice Guideline for Anemia in Chronic Kidney Disease. 2016; time of surgery. Ann Oncol. 2011;22(9):1988‐1998.
https​: //kdigo.org/wp-conte​n t/uploa​d s/2016/10/KDIGO-2012- 29. Thomas G, Ali S, Hoebers FJ, et al. Phase III trial to evaluate the
Anemia-Guide​line-Engli​sh.pdf. Accessed June 19, 2019. efficacy of maintaining hemoglobin levels above 12.0 g/dL with
11. Bohlius J. Management of cancer associated anemia with erythropoi‐ erythropoietin vs above 10.0 g/dL without erythropoietin in ane‐
esis‐stimulating agents: ASCO/ASH clinical guideline update. 2019; mic patients receiving concurrent radiation and cisplatin for cervi‐
http://www.blood ​ a dvan ​ c es.org/conte ​ n t/3/8/1197?sso-check​ cal cancer. Gynecol Oncol. 2008;108(2):317‐325.
ed=true. Accessed June 19, 2019. 30. Tonia T, Mettler A, Robert N, et al. Erythropoietin or darbe‐
12. Aapro M, Jelkmann W, Constantinescu SN, Leyland‐Jones B. poetin for patients with cancer. Cochrane Database Syst Rev.
Effects of erythropoietin receptors and erythropoiesis‐stim‐ 2012;12:CD003407.
ulating agents on disease progression in cancer. Br J Cancer. 31. Grant MD, Piper M, Bohlius J, et al. Epoetin and Darbepoetin for
2012;106(7):1249‐1258. Managing Anemia in Patients Undergoing Cancer Treatment: com-
13. Bittorf T, Buchse T, Sasse T, Jaster R, Brock J. Activation of the parative Effectiveness Update. Report No:13-EHC077-EF AHRQ
transcription factor NF‐kappaB by the erythropoietin receptor: Comparative Effectiveness Reviews. Rockville, MD: Agency for
structural requirements and biological significance. Cell Signal. Healthcare Research and Quality (US); 2013.
2001;13(9):673‐681. 32. Tonelli M, Hemmelgarn B, Reiman T, et al. Benefits and harms of
14. Saintigny P, Besse B, Callard P, et al. Erythropoietin and erythropoi‐ erythropoiesis‐stimulating agents for anemia related to cancer: a
etin receptor coexpression is associated with poor survival in stage I meta‐analysis. CMAJ. 2009;180(11):E62‐71.
non‐small cell lung cancer. Clin Cancer Res. 2007;13(16):4825‐4831. 33. Glaspy J, Crawford J, Vansteenkiste J, et al. Erythropoiesis‐stim‐
15. Acs G, Acs P, Beckwith SM, et al. Erythropoietin and eryth‐ ulating agents in oncology: a study‐level meta‐analysis of survival
ropoietin receptor expression in human cancer. Can Res. and other safety outcomes. Br J Cancer. 2010;102(2):301‐315.
2001;61(9):3561‐3565. 34. Aapro M, Osterwalder B, Scherhag A, Burger HU. Epoetin‐beta
16. Arcasoy MO, Jiang X, Haroon ZA. Expression of erythropoietin re‐ treatment in patients with cancer chemotherapy‐induced anaemia:
ceptor splice variants in human cancer. Biochem Biophys Res Comm. the impact of initial haemoglobin and target haemoglobin levels
2003;307(4):999‐1007. on survival, tumour progression and thromboembolic events. Br J
17. Ribatti D, Marzullo A, Nico B, Crivellato E, Ria R, Vacca A. Cancer. 2009;101(12):1961‐1971.
Erythropoietin as an angiogenic factor in gastric carcinoma. 35. Ludwig H, Crawford J, Österborg A, et al. Pooled analysis of in‐
Histopathology. 2003;42(3):246‐250. dividual patient‐level data from all randomized, double‐blind,
18. Lai SY, Childs EE, Xi S, et al. Erythropoietin‐mediated activation of placebo‐controlled trials of darbepoetin alfa in the treatment
JAK‐STAT signaling contributes to cellular invasion in head and neck of patients with chemotherapy‐induced anemia. J Clin Oncol.
squamous cell carcinoma. Oncogene. 2005;24(27):4442‐4449. 2009;27(17):2838‐2847.
19. Yasuda Y, Masuda S, Chikuma M, Inoue K, Nagao M, Sasaki 36. Pirker R, Ramlau RA, Schuette W, et al. Safety and efficacy of dar‐
R. Estrogen‐dependent production of erythropoietin in bepoetin alpha in previously untreated extensive‐stage small‐cell
uterus and its implication in uterine angiogenesis. J Biol Chem. lung cancer treated with platinum plus etoposide. J Clin Oncol.
1998;273(39):25381‐25387. 2008;26(14):2342‐2349.
LATCHA | 519

37. Bohlius J, Tonia T, Schwarzer G. Twist and shout: one decade of 55. Timp JF, Braekkan SK, Versteeg HH, Cannegieter SC.
meta‐analyses of erythropoiesis‐stimulating agents in cancer pa‐ Epidemiology of cancer‐associated venous thrombosis. Blood.
tients. Acta Haematol. 2011;125(1–2):55‐67. 2013;122(10):1712‐1723.
38. Koulouridis I, Alfayez M, Trikalinos TA, Balk EM, Jaber BL. Dose of 56. Wun T, White RH. Epidemiology of cancer‐related venous throm‐
erythropoiesis‐stimulating agents and adverse outcomes in CKD: a boembolism. Best Pract Res Clin Haematol. 2009;22(1):9‐23.
metaregression analysis. Am J Kidney Dis. 2013;61(1):44‐56. 57. Chew HK, Wun T, Harvey D, Zhou H, White RH. Incidence of ve‐
39. Szczech LA, Barnhart HX, Inrig JK, et al. Secondary analysis of the nous thromboembolism and its effect on survival among patients
CHOIR trial epoetin‐alpha dose and achieved hemoglobin out‐ with common cancers. Arch Intern Med. 2006;166(4):458‐464.
comes. Kidney Int. 2008;74(6):791‐798. 58. Blom JW, Doggen CJ, Osanto S, Rosendaal FR. Malignancies, pro‐
40. Solomon SD, Uno H, Lewis EF, et al. Erythropoietic response and thrombotic mutations, and the risk of venous thrombosis. JAMA.
outcomes in kidney disease and type 2 diabetes. N Engl J Med. 2005;293(6):715‐722.
2010;363(12):1146‐1155. 59. Khorana AA. Venous thromboembolism prevention in cancer out‐
41. System USRD. 2018 Annual Data Report. 2018; https​ ://www. patients. J Natl Compr Canc Netw. 2013;11(11):1431‐1438.
usrds.org/adr.aspx. Accessed July 23, 2019. 60. Key NS, Khorana AA, Kuderer NM, et al. Venous thromboembolism
42. Wachterman MW, O’Hare AM, Rahman O‐K, et al. One‐year mor‐ prophylaxis and treatment in patients with cancer: ASCO clinical
tality after dialysis initiation among older adults. JAMA Intern Med. practice guideline update. J Clin Oncol. 2019;37:JCO1901461.
2019;179(7):987. 61. mdcalc.com. Khorana Risk Score for Venous Thromboembolism
43. Cancer Screening Guidelines. https​://www.uspre​venti​veser ​vices​ in Cancer Patients. 2019; https​ ://www.mdcalc.com/khora​ na-
taskf​o rce.org/Page/Docum ​e nt/Updat​e Summ​a ryFi​n al/cervi​c al- risk-score-venous-throm​boemb​olism-cancer-patients. Accessed
cancer-scree​ning. Accessed September 19, 2019. August 16, 2019.
44. Couchoud C, Labeeuw M, Moranne O, et al. A clinical score to pre‐ 62. Hellstrom‐Lindberg E, Gulbrandsen N, Lindberg G, et al. A
dict 6‐month prognosis in elderly patients starting dialysis for end‐ validated decision model for treating the anaemia of myelo‐
stage renal disease. Nephrol Dial Transplant. 2009;24(5):1553‐1561. dysplastic syndromes with erythropoietin + granulocyte col‐
45. Zou S, Stramer SL, Dodd RY. Donor testing and risk: current ony‐stimulating factor: significant effects on quality of life. Br J
prevalence, incidence, and residual risk of transfusion‐trans‐ Haematol. 2003;120(6):1037‐1046.
missible agents in US allogeneic donations. Transfus Med Rev. 63. Littlewood TJ, Zagari M, Pallister C, Perkins A. Baseline and early
2012;26(2):119‐128. treatment factors are not clinically useful for predicting individual
46. Stramer SL, Notari EP, Krysztof DE, Dodd RY. Hepatitis B virus test‐ response to erythropoietin in anemic cancer patients. Oncologist.
ing by minipool nucleic acid testing: does it improve blood safety? 2003;8(1):99‐107.
Transfusion. 2013;53(10 Pt 2):2449‐2458. 64. Locatelli F, Fishbane S, Block GA, Macdougall IC. Targeting hypoxia‐
47. FDA. Approved cellular and gene therapy products: Yescarta (axi‐ inducible factors for the treatment of anemia in chronic kidney dis‐
cabtagene ciloleucel). ease patients. Am J Nephrol. 2017;45(3):187‐199.
48. FDA. Approved cellular and gene therapy products: Kymriah 65. Fallah J, Rini BI. HIF inhibitors: status of current clinical develop‐
(tisagenlecleucel). ment. Curr Oncol Rep. 2019;21(1):6.
49. Yabu JM, Anderson MW, Kim D, et al. Sensitization from transfu‐ 66. Koh MY, Powis G. Passing the baton: the HIF switch. Trends Biochem
sion in patients awaiting primary kidney transplant. Nephrol Dial Sci. 2012;37(9):364‐372.
Transplant. 2013;28(11):2908‐2918. 67. Albadari N, Deng S, Li W. The transcriptional factors HIF‐1 and
50. Leffell MS, Kim D, Vega RM, et al. Red blood cell transfusions and HIF‐2 and their novel inhibitors in cancer therapy. Expert Opin Drug
the risk of allosensitization in patients awaiting primary kidney Discov. 2019;14(7):667‐682.
transplantation. Transplantation. 2014;97(5):525‐533. 68. Chiu D‐C, Tse A‐W, Xu I‐J, et al. Hypoxia inducible factor HIF‐1
51. Singer ST, Wu V, Mignacca R, Kuypers FA, Morel P, Vichinsky EP. promotes myeloid‐derived suppressor cells accumulation through
Alloimmunization and erythrocyte autoimmunization in transfu‐ ENTPD2/CD39L1 in hepatocellular carcinoma. Nat Commun.
sion‐dependent thalassemia patients of predominantly asian de‐ 2017;8(1):517.
scent. Blood. 2000;96(10):3369‐3373. 69. Semenza GL. Defining the role of hypoxia‐inducible factor 1 in can‐
52. van de Watering L, Hermans J, Witvliet M, Versteegh M, Brand A. cer biology and therapeutics. Oncogene. 2010;29(5):625‐634.
HLA and RBC immunization after filtered and buffy coat‐depleted 70. Artunc F, Risler T. Serum erythropoietin concentrations and re‐
blood transfusion in cardiac surgery: a randomized controlled trial. sponses to anaemia in patients with or without chronic kidney dis‐
Transfusion. 2003;43(6):765‐771. ease. Nephrol Dial Transplant. 2007;22(10):2900‐2908.
53. Zwald F, Leitenberger J, Zeitouni N, et al. Recommendations for
solid organ transplantation for transplant candidates with a pre‐
transplant diagnosis of cutaneous squamous cell carcinoma, Merkel
How to cite this article: Latcha S. Anemia management in
cell carcinoma and melanoma: a consensus opinion from the
cancer patients with chronic kidney disease. Semin Dial.
International Transplant Skin Cancer Collaborative (ITSCC). Am J
Transplant. 2016;16(2):407‐413. 2019;32:513–519. https​://doi.org/10.1111/sdi.12841​
54. Chapman JR, Webster AC, Wong G. Wong G. Cancer in the trans‐
plant recipient. Cold Spring Harb Perspect Med. 2013;3(7):a015677.