6348 Full

Published OnlineFirst September 4, 2012; DOI: 10.1158/1078-0432.
CCR-12-1397
Clinical
Cancer
Imaging, Diagnosis, Prognosis Research
Serum Lactate Dehydrogenase Is Prognostic for Survival in

Patients with Bone Metastases from Breast Cancer: A
Retrospective Analysis in Bisphosphonate-Treated Patients
Janet E. Brown1,2, Richard J. Cook3, Allan Lipton4, and Robert E. Coleman2
Abstract
Purpose: Survival is highly variable in women with bone metastases from breast cancer and prognostic
factors are needed. We analyzed data from a phase III trial comparing zoledronic acid (ZOL) with pamidro-
nate in patients with breast cancer and bone metastases to identify variables prognostic for overall survival.
Experimental Design: Patients who received ZOL (n ¼ 435) with bone marker assessments and complete
baseline data were included. Relative risks (RR) of death over 24 months were assessed using a stratified Cox
regression analysis. A reduced model was generated using stepwise backward elimination until only
significant (P < 0.05) variables remained.
Results: Only 5 of 19 variables analyzed remained significantly prognostic for survival in the reduced
multivariate model. These included age more than 50 years (RR 1.78–2.53, P 0.01 for each decade >50
versus 50); Functional Assessment of Cancer Therapy-General (FACT-G) score less than 65 units (P < 0.05
vs. 75 units); impaired (PS 1) versus fully active (PS ¼ 0) Eastern Cooperative Oncology Group (ECOG)
performance status (RR 1.74, P < 0.01); prior versus no prior chemotherapy (RR 1.97; P < 0.01), and lactate
dehydrogenase (LDH) levels. Lactate dehydrogenase upper limit of normal (ULN) but < 2 ULN
correlated with a two-fold increased risk of death, and LDH > 2 ULN correlated with a six-fold increased
risk of death versus LDH < ULN (P < 0.0001 for both). Baseline bone marker levels were not significantly
correlated with survival after adjustment for other significant covariates.
Conclusions: This retrospective analysis shows that LDH levels correlate strongly with survival in patients
with bone metastases from breast cancer and confirms the relevance of previously described prognostic
factors. Clin Cancer Res; 18(22); 6348–55. 2012 AACR.
Introduction cemia of malignancy. Overall, SREs have been associated with

Among patients with advanced breast cancer, approximate- reduced performance status, impaired health-related quality
ly 65% to 75% will develop bone metastases (1), which of life, shortened survival, and increased bone pain (2, 3).
weaken the structural integrity of the skeleton and can result There is ongoing interest in identifying prognostic vari-
in painful and potentially disabling skeletal-related events ables that could predict a patient’s risk for disease progression
(SRE) including pathologic fracture, the need for radiotherapy and death. For example, age, clinical status of nodes, tumor
or surgery to bone, spinal cord compression, and hypercal- size, race, and number of positive nodes have been reported
as prognostic factors for survival in patients with breast cancer
that is limited to local or regional sites (4). More recently,
Authors' Affiliations: 1Cancer Research UK Experimental Cancer Med- factors such as human epidermal growth factor receptor-2
icine Centres at Leeds and Sheffield; 2Cancer Research Centre, Academic
Unit of Clinical Oncology, Weston Park Hospital, Sheffield; 3University of (HER2) expression have been strongly associated with overall
Waterloo, Waterloo, Ontario, Canada; and 4Milton S. Hershey Medical survival (OS) prognosis (5), and there has been considerable
Center, Penn State University, Hershey, Pennsylvania progress in the development of gene-expression–based prog-
Note: Abstract and poster presentations include the following conferences: nosis for survival (6). In addition, the identification of circu-
EBCC 2008, CIBD 2008, and ASCO-BC 2008. Table 2, Fig. 1, and Fig. 2 lating tumor cells (CTC) has been approved as a prognostic
were presented in a poster: Coleman RE, Brown JE, Cook R. Serum lactate
dehydrogenase (LDH) is a significant prognostic variable for survival in marker to assess response to treatment in patients with
patients with metastatic breast cancer: a multivariate analysis. Presented metastatic breast cancer (7), and the CellSearch (Quest Diag-
at: EBCC 6: European Breast Cancer Conference; 2008 April 15–19; Berlin,
Germany. Abstract nr 468.
nostics) assay for CTCs is claimed to be a strong, independent
predictor of OS, and progression-free survival. Among
Corresponding Author: Janet E. Brown, Cancer Research UK Clinical
Centre, Leeds, JIF Building, St. James' University Hospital, Becket Street, women with bone metastases from breast cancer, median
Leeds LS9 7TF, United Kingdom. Phone: 44-113-206-4184; Fax: 44-113- survival is approximately 2 to 3 years (8), in comparison with
246-0136; E-mail: [email protected] the 12- to 18-month median survival in patients with liver
doi: 10.1158/1078-0432.CCR-12-1397 metastases (9). Unfortunately, specific prognostic models
2012 American Association for Cancer Research. for long-term survival in patients with bone metastases from
6348 Clin Cancer Res; 18(22) November 15, 2012
Downloaded from clincancerres.aacrjournals.org on February 21, 2021. © 2012 American Association for Cancer Research.
Published OnlineFirst September 4, 2012; DOI: 10.1158/1078-0432.CCR-12-1397
Prognostic Value of LDH in Patients with Bone Metastases
III trial comparing ZOL with pamidronate (21–23). All

Translational Relevance
patients had at least 1 radiographically confirmed bone
Survival is highly variable in women with bone meta- lesion, Eastern Cooperative Oncology Group performance
stases from breast cancer, and prognostic factors are status (ECOG PS) of 0 to 2, serum creatinine 3 mg/dL (265
needed. This retrospective exploratory analysis assess- mmol/L) or more, and no prior bisphosphonate therapy. In
ed data from a phase III trial comparing zoledronic acid this trial, patients were randomized to pamidronate (90 mg
with pamidronate in patients with breast cancer and bone every 3–4 weeks) or either of 2 ZOL treatment arms (4 mg or
metastases to identify variables prognostic for overall 8 mg every 3–4 weeks) for up to 2 years. During the trial, the
survival. Results from this analysis show that lactate ZOL 8-mg dose was reduced to 4 mg for renal safety and is
dehydrogenase (LDH) levels correlate strongly with sur- referred to as the 8/4-mg group. All patients enrolled in the
vival in patients with bone metastases from breast cancer. trial provided written informed consent. To preserve homo-
Furthermore, this analysis also confirms the relevance geneity of bisphosphonate treatment and because of the
of previously described prognostic factors such as age, large numbers of patients available, only patients with
Eastern Cooperative Oncology Group performance breast cancer in the ZOL treatment groups were included
status, prior chemotherapy, and Functional Assessment in the analyses reported herein. In the subset of patients
of Cancer Therapy-General (FACT-G) score. On the who enrolled in North American treatment centers, all
basis of the current study, the combination of LDH with baseline prognostic factors including biochemical markers
other prognostic factors into a weighted prognostic of bone turnover (e.g., BSAP and NTX) were assessed at
score for survival in patients with bone metastases from baseline and at months 1, 3, 6, and 12 of the study.
breast cancer would be a valuable next step.
Laboratory procedures
Laboratory samples were collected and processed locally
on study enrollment at the baseline visit according to the
breast cancer are not well developed (10–14). We have strict standard operating procedures (SOP) set by the study.
previously shown that elevated baseline levels of the bio- Sample storage and shipping to regional GCP-accredited
chemical marker N-telopeptide of type I collagen (NTX) were central laboratories (Mayo Medical Laboratories in North
associated with an increased risk of SREs (15, 16), and and South America; BARC in Europe, Australia, and South
normalization of NTX levels correlated with reduced risks Africa) also adhered to the study SOPs. Samples were
of fracture, SREs, and death versus persistently elevated NTX analyzed as per guidelines set by each regional central
after 3 months of treatment with zoledronic acid (ZOL; laboratory. Patient baseline samples for laboratory analysis
refs. 17, 18). Factors such as prior SRE, number of bone were measured at the time of enrollment, not at the time of
lesions, and the severity of bone pain may also provide bone metastasis diagnosis.
insight about SRE risk (19, 20). Because a range of systemic
treatments is now available and many patients with breast Statistical analyses
cancer will survive for several years after diagnosis of bone Patients with breast cancer who received ZOL (both the 4-
metastases, it is important to identify factors that are prog- mg and 8/4-mg groups) who also had assessments of
nostic for survival to tailor treatments accordingly. biochemical markers of bone metabolism and complete
The phase III, randomized, controlled trial comparing data sets of baseline variables (n ¼ 435) were included in the
ZOL with pamidronate in patients with bone metastases statistical models. Nineteen baseline variables with poten-
from breast cancer (21, 22) generated an extensive database tial prognostic value, including age, race, weight, predom-
of baseline clinical and biochemical parameters with long- inant radiographic appearance of bone lesion(s)—osteoly-
term (25-month) clinical follow-up that can be used for tic, osteoblastic, and other—Functional Assessment of
correlative analyses (23). A recent multivariate analysis of Cancer Therapy-General (FACT-G) score, ECOG PS, history
baseline variables in this trial identified prognostic factors of SRE, history of chemotherapy, concomitant endocrine
for SRE risk including age, pain score, prior history of an therapy, presence of extraskeletal metastases (with this
SRE, predominant lesion type, elevated bone-specific alka- variable being subdivided into liver and nonliver metastases
line phosphatase (BSAP), and elevated lactate dehydroge- as it was possible that patients with liver metastases may
nase (LDH; ref. 24). The current exploratory analysis uses have a different survival pattern compared with those with
data collected prospectively in the above trial and was other extraskeletal metastases), NTX, BSAP, lymphocyte
conducted to investigate prognostic variables for OS among count, hemoglobin, serum glutamic oxaloacetic acid, albu-
patients with bone metastases from breast cancer enrolled min, creatinine, and total serum LDH, were included in the
in the ZOL arm of the pamidronate-controlled trial. univariate and full multivariate Cox regression models to
assess relative risk and associated 95% confidence intervals
Materials and Methods (CI) for death throughout the 24-month trial.
Patient population All models were stratified based on time since diagnosis
Data were obtained for patients with bone metastases of primary cancer (<5 vs. 5 years) and by ZOL treatment
from breast cancer who were enrolled (October 1998 to group. Patient age was categorized into <50, 50 to <60, 60
December 1999) in the international, randomized, phase to <70, and 70 years of age. Patient weight was categorized
www.aacrjournals.org Clin Cancer Res; 18(22) November 15, 2012 6349
Brown et al.
into <60, 60 to <70, 70 to <80, and 80 kilograms.

Table 1. Baseline demographics and laboratory
Patient FACT-G score was categorized into <65, 65 to
values
<75, 75 to <90, and 90 units. At the time this trial was
conducted, HER2 and estrogen receptor status of the primary Patients,
tumor were not routinely assessed. Laboratory values (includ- Variable (evaluable patients, n) n (%)
ing bone marker levels) were treated as dichotomous vari- Age, y N ¼ 742
ables using either the established upper limit of normal 0 to <50 207 (28)
(ULN) as the cut-off point or the respective ULN from each 50 but <60 213 (29)
specific laboratory for each patient’s assessments; the only 60 but <70 181 (24)
exception was for hemoglobin, where the median value of 70 141 (19)
12 g/dL was used. Serum LDH levels were categorized into 3 Duration of cancer at study entry, y N ¼ 742
different levels using laboratory specific (because of variations <5 375 (51)
in this international trial, absolute cut-off values would have 5 367 (49)
been less appropriate) ULN and 2 ULN as cut-off points ECOG PS N ¼ 740
(i.e., LDH < ULN; ULN LDH < 2 ULN; LDH 2 ULN). Fully active (PS ¼ 0) 256 (35)
All other variables were treated as dichotomous (yes vs. no). Some impairment (PS 1) 484 (65)
For all analyses, associations were considered statistically Prior SRE N ¼ 740
significant if their associated P value was <0.05. Reduced No 302 (41)
Yes 438 (59)
multivariate models were generated by stepwise backward
History of chemotherapy N ¼ 734
elimination until only significant (P < 0.05) variables
No 179 (24)
remained. Kaplan–Meier estimates were computed to
Yes 555 (76)
graphically summarize the distribution of time to death
NTX, nmol/mmol creatinine N ¼ 490
stratified by newly identified prognostic variables.
<64 197 (40)
Results 64 293 (60)
BSAP, U/L N ¼ 501
Patients
<146 130 (26)
Among patients with bone metastases from breast cancer 146 371 (74)
enrolled in the phase III trial comparing ZOL with pami- Hemoglobin, g/dL N ¼ 729
dronate (N ¼ 1,130), a total of 742 patients were random- 12 364 (50)
ized to the 2 ZOL arms of the study (ref. 22; Table 1; ref. 24). >12 365 (50)
The majority of patients (81%) were less than 70 years of AST, U/L N ¼ 735
age, had some impairment in ECOG-PS (65%), had a prior <ULN 446 (61)
SRE (59%), and had received prior chemotherapy (76%). ULN 289 (39)
Among the 742 patients in the ZOL arms, a complete Albumin, g/L N ¼ 735
baseline assessment for all 19 variables of interest was 35 94 (13)
available for 435 patients, who were therefore included in >35 641 (87)
the models described below. The median time from diag- Creatinine, mg/dL N ¼ 735
nosis of bone metastasis to study entry was 3.75 months. <ULN 387 (53)
ULN 348 (47)
Univariate and multivariate models of overall survival LDH, U/L N ¼ 735
Risk factors that could potentially influence OS were <ULN 537 (73)
evaluated in univariate analyses for each of the 19 baseline ULN 198 (27)
quartile-based categorical and dichotomous variables.
NOTE: Adapted with kind permission from Springer
Among these variables, 15 were found to be associated with
ScienceþBusiness Media: Prognostic factors for skeletal
a significant increase in the risk of death, including low
complications from metastatic bone disease in breast
weight, low FACT-G score, ECOG PS 1, prior SRE, prior
cancer. 123, 2010, page 770, Brown JE, Lipton A, Costa L,
chemotherapy, concomitant hormone therapy, metastases,
and Coleman R. Table 1.
osteoblastic lesions, elevated NTX, elevated BSAP, elevated
hemoglobin, elevated aspartate aminotransferase (AST),
elevated albumin, and elevated LDH (Table 2). The remain-
ing variables (age, race, lymphocyte counts, and serum being the only statistically significant laboratory parameter.
creatinine) were not significantly associated with increased The 5 variables that were associated with a statistically
risk of death. significant increase in risk of death remained significant
However, only 5 covariates were significantly associated covariates after applying stepwise backward elimination
with increased risk of death after adjustment for all other of all nonsignificant variables in the multivariate model
variables in the full multivariate model. These included (Fig. 1). In the reduced model, elevated LDH [a factor
advanced age, low FACT-G score, ECOG PS 1, prior previously identified as a prognostic marker in other cancers
chemotherapy, and elevated LDH (Table 2), with LDH (25–29), but not specifically in patients with advanced
6350 Clin Cancer Res; 18(22) November 15, 2012 Clinical Cancer Research
Table 2. Univariate and multivariate models for overall survival in patients with bone metastases from
breast cancer
Univariate Full multivariate
Factor RR 95% CI P RR 95% CI P

Versus age < 50 years
50 y age < 60 y 1.359 0.932–1.981 0.1110 1.795 1.174–2.746 0.0069
60 y age < 70 y 1.122 0.753–1.674 0.5710 1.598 1.010–2.526 0.0450
70 y age 1.494 0.996–2.241 0.0523 2.472 1.533–3.988 0.0002
Global test 0.1857 0.0020
Versus weight 80 kg
Weight < 60 kg 1.816 1.243–2.654 0.0020 1.590 1.044–2.420 0.0306
60 kg weight < 70 kg 0.929 0.636–1.356 0.7020 0.934 0.625–1.394 0.7379
70 kg weight < 80 kg 0.922 0.614–1.384 0.6941 1.022 0.663–1.576 0.9205
Versus FACT-G < 65 units
65 units FACT-G < 75 units 0.713 0.483–1.053 0.0894 0.700 0.457–1.073 0.1017
75 units FACT-G < 90 units 0.645 0.452–0.920 0.0154 0.606 0.410–0.895 0.0119
90 units FACT-G 0.442 0.296–0.659 <0.0001 0.540 0.344–0.845 0.0071
ECOG PS 1 vs. 0 2.216 1.608–3.052 <0.0001 1.572 1.097–2.253 0.0137
Versus white race
Black/Asian/other 1.152 0.778–1.705 0.4809 0.924 0.602–1.418 0.7170
1 Prior SRE vs. 0 1.396 1.044–1.868 0.0246 1.130 0.812–1.572 0.4677
Prior chemotherapy (1 vs. 0) 1.779 1.223–2.586 0.0026 1.812 1.169–2.808 0.0078
Concomitant HT (yes vs. no) 0.600 0.449–0.801 0.0005 0.938 0.673–1.308 0.7063
Metastases sites
1 liver vs. 0 on liver 1.760 1.288–2.404 0.0004 1.153 0.819–1.624 0.4148
1 lung/brain/other vs. 0 1.463 1.103–1.940 0.0082 1.272 0.930–1.741 0.1321
Predominant lesion (vs. osteolytic)
Osteoblastic 0.679 0.499–0.923 0.0136 0.794 0.575–1.096 0.1599
Other 0.690 0.476–1.000 0.0500 0.659 0.447–0.970 0.0344
NTX 64 vs. < 64 nmol/mmol creatinine 1.627 1.213–2.183 0.0012 1.151 0.827–1.603 0.4055
BSAP 146 vs. < 146 U/L 1.596 1.135–2.244 0.0072 1.084 0.733–1.602 0.6866
Lymphocytes ULN vs. < ULN 0.649 0.341–1.237 0.1890 0.967 0.489–1.913 0.9225
Hemoglobin 12 vs. < 12 g/dL 0.678 0.512–0.897 0.0066 0.901 0.659–1.233 0.5148
AST ULN vs. < ULN 1.931 1.466–2.543 <0.0001 1.212 0.852–1.725 0.2846
Serum albumin 35 vs. < 35 g/L 0.460 0.316–0.670 <0.0001 0.780 0.503–1.212 0.2693
Serum creatinine ULN vs. < ULN 0.884 0.659–1.185 0.4087 1.092 0.787–1.515 0.5994
Versus LDH < ULN
ULN LDH < 2 ULN 2.190 1.593–3.011 <0.0001 1.563 1.042–2.344 0.0307
2 ULN LDH 5.188 2.965–9.077 <0.0001 4.503 2.269–8.935 <0.0001
Global test <0.0001 <0.0001
Abbreviations: HT, hormone therapy; RR, risk ratio.
breast cancer and bone metastases] showed a strong corre- between-group differences were apparent throughout the
lation with risk of death. Patients with LDH levels 2 course of the study (P < 0.0001; Fig. 2). Patients with LDH
ULN had an almost 6-fold increased risk of death compared levels < ULN had the best OS, and approximately 65% of
with patients whose LDH levels were below the ULN (RR patients were still alive at 24 months’ follow-up. In contrast,
5.937; P < 0.0001). the median survival among patients with LDH levels ULN
but <2 ULN was approximately 17 months, and those
LDH as a prognostic marker of survival in advanced patients with LDH levels 2 ULN had the worst OS
breast cancer (median 5 months). At the 12-month time point, Cox
Using the ULN and 2 ULN as cutoff points, the resulting regression analyses showed that significant prognostic fac-
3 LDH populations had significantly different survival, and tors predictive of >30% risk of death included high LDH,
Brown et al.
Figure 1. Forest plot of significant prognostic variables in reduced multivariate Cox model for death in patients with bone metastases from breast cancer.
Horizontal lines represent the 95% confidence interval. Global P values are for all categories versus the reference category and are derived from the
full Cox multivariate analysis.
weight < 60 kg, albumin < 35 g/L, FACT-G < 65 units, and Other prognostic variables that remained significant in
SGOT ULN. Notably, baseline LDH levels were a consis- multivariate models included age, ECOG performance sta-
tent predictor of mortality throughout the study (e.g., at the tus, prior chemotherapy, and FACT-G score. Because these
6-, 12-, and 18-month time points). variables have been well recognized as factors prognostic for
survival in patients with metastatic breast cancer (14, 31),
they provide validation for the methodology used in the
Discussion current analyses. Univariate and multivariate analyses have
The registration trial comparing ZOL with pamidronate identified baseline bone biomarker levels such as NTX and
in patients with bone metastases from breast cancer BSAP as prognostic factors for occurrence of SREs in patients
generated an extensive database of demographic charac- with bone metastases (24, 32). However, although both
teristics, concomitant therapy, and clinical outcomes, NTX and BSAP were significantly correlated with risk of
including survival and laboratory parameters that provide death in the univariate model in the current analysis, this
a rich data source for exploratory analyses. All patients in statistically significant correlation was not maintained for
the current analysis received ZOL therapy, which remains either biomarker in the multivariate model (Table 2).
a standard of care for patients with bone metastases from Notably, LDH is a key enzyme in energy production in
solid tumors (25). Access to this database has permitted many tissues and cell types, and LDH serum concentrations
this large, multicenter, multivariate analysis of factors become elevated following tissue injury or during disease
prognostic for survival in patients with bone metastases states. The potential role of LDH as a prognostic biomarker
from breast cancer (30). A key finding from these analyses in oncology has long been recognized. In patients with
is the emergence of LDH as a significant and powerful Ewing’s sarcoma of bone (N ¼ 618), baseline serum LDH
prognostic variable for risk of death in patients with levels were significantly higher in patients with metastatic
advanced breast cancer. disease (P < 0.0001) and prognostic of shorter disease-free
patients (n ¼ 299) with metastatic cancer treated at a single

center were included, of whom 132 patients had breast
cancer (estimated 100–110 with bone metastases). Few
previous studies have focused on the prognostic value of
LDH in patients with bone metastases. In a retrospective
analysis of patients with bone metastases from castration-
resistant prostate cancer (N ¼ 643), elevated baseline serum
LDH levels (>454 U/L) were associated with a nearly 3-fold
increased risk of death in the univariate model (RR 2.83; P <
0.001), and with a 2-fold increased risk of death in the
reduced multivariate model (RR 2.01; P < 0.001; ref. 43).
Analyses in a smaller group of patients with bone metastases
from prostate cancer (N ¼ 60) showed that serum LDH level
was a significant prognostic indicator for survival both in
univariate (P ¼ 0.0049) and multivariate (P ¼ 0.0371)
Figure 2. Kaplan–Meier estimates for overall survival of patients with bone analyses (44). The current study is believed to be the first
metastases from breast cancer treated with ZOL using multivariate to focus on a large cohort of breast cancer patients with bone
modeling by LDH levels. metastases) in a multicenter setting. Although it is recog-
nized that different isoforms of LDH exist and can be
survival in patients with localized disease versus patients measured separately, a key advantage of total serum LDH
with normal LDH levels (P < 0.0001; ref. 33). In small cell as a prognostic indicator is that it is extremely convenient
lung cancer, serum LDH has been shown to be a significant and inexpensive to measure. It is fortunate that total serum
predictor of survival (29, 34) and is used routinely for LDH in this and other oncology studies is strongly associ-
assisting treatment decisions using, for example, the Man- ated with survival, without the need to measure individual
chester Prognostic Score, which includes serum LDH isoforms.
among a panel of markers (35). In non–small cell lung The current exploratory analysis used a patient popula-
cancer, analysis of a large database (N ¼ 2,531) found that tion with bone metastases treated with ZOL, which remains
normal baseline LDH levels were predictive of better sur- a current standard of care for bone-targeted therapy. How-
vival (25). Correlations between baseline LDH levels and ever, after the data were collected in this patient population,
survival also have been observed in patients with multiple estrogen receptor and HER2 status in breast cancer came to
myeloma, mantle cell lymphoma, and prostate cancer (26– be routinely measured on tissue blocks. Furthermore, there
28). Recently, studies in metastatic melanoma have have been advances in cytotoxic therapy with newer agents
reported that LDH is prognostic for survival (36) and LDH such as docetaxel, routine use of aromatase inhibitors as a
correlated significantly with stage of disease (37). In other key component of hormonal therapy, and routine use of
recent studies, serum LDH levels were the main prognostic targeted therapies such as trastuzumab. In addition, CTCs
factor for colorectal cancer in a survival prognostic model are in more common use as potential prognostic markers.
(38). Moreover, the addition of LDH to the International Although changes inevitably occur during the gathering of
Myeloma Staging Model was found to improve prognostic data from large studies, the question arises as to whether
value (39). Taken together, these various analyses support such changes will affect the prognostic relevance of LDH
the potential prognostic value of baseline LDH levels across determined in this study.
several tumor types. Indirect evidence is reassuring in this regard. For example,
Construction and validation of a practical prognostic in a recent study of potential prognostic factors in metastatic
index for patients with metastatic breast cancer was reported prostate cancer (n ¼ 104), LDH remained a significant
by Yamamoto and colleagues (40). Multivariate analyses prognostic factor even though newer factors such as HER2
of data from 233 patients with metastatic breast cancer overexpression were included in a multivariate analysis
showed that LDH was among 5 factors prognostic for (45). Also, in castration-resistant prostate cancer, a study
survival. These 5 factors were used to construct a prognostic of the prognostic value of CTCs concluded that the factors
index that was prospectively evaluated in a validation set most predictive for survival were high LDH concentration
of 315 patients with metastatic breast cancer. Notably, a and elevated CTC counts (P < 0.001, both; ref. 46). In a
new prognostic score that includes serum LDH measure- study of 996 myeloma patients, Terpos and colleagues
ments has been proposed for patients with brain meta- found that high versus normal LDH had a major impact
stases. This 7-factor prognostic score, which includes serum on OS even when patients have received novel agent-based
LDH and HER2 overexpression, successfully predicted sur- therapies (median OS 21 versus 51 months, respectively; P <
vival for a heterogeneous group of 130 patients (41). 0.001; ref. 39).
A recent study claimed to prospectively validate 2 prog- We have considered the possible effects of "false posi-
nostic scores (combining LDH and either performance tives" that have been reported to occur in assessment of
status or lymphocyte count) for survival in patients with LDH. The effects of any bias introduced by these have been
cancer after first-line chemotherapy (42). All eligible minimized by using categorical LDH values instead of
Brown et al.
absolute measurements. Also, the possibility of false pos- be incorporated into future large breast cancer studies, and
itive LDH results would actually reduce the chance of LDH also may be an important stratification factor in the
showing an association when one exists. Thus, if false design of future randomized trials.
positives were present in the data, the model would more
likely yield conservative estimates of the association Disclosure of Potential Conflicts of Interest
between LDH and survival. J. Brown has received honoraria from speakers bureau from Novartis
There are potential limitations in this study worthy of and Amgen and is a consultant and/or advisory board member for
Novartis, Amgen, Roche, and Bristol-Myers Squibb. R. Cook is a consul-
note. Laboratory-specific ULNs for LDH were used to tant/advisory board member for Abbott, Amgen, and Novartis. A. Lipton
specify cut-off points, in preference to an absolute LDH has received commercial research grants from Novartis, Monogram Bios-
value as the cutoff. This was considered appropriate to ciences, and Oncogene Science; has received honoraria from speakers
bureau from Novartis, Amgen, and Genentech; is a consultant/advisory
reflect local factors and because some variation in abso- board member for Novartis, Amgen, Galapagos, and Acceleron Pharm;
lute reference ranges was reflected in the ULN across and has given expert testimony for Novartis. R. Coleman has received
other commercial support from Novartis; honoraria from speakers bureau
centers. Consequently, it is difficult to directly compare from Novartis and Amgen; is a consultant/advisory board member for
data obtained in this study with those previously reported Novartis and Amgen; and has previously given expert testimony on the
for patients with advanced cancer with or without bone behalf of Novartis.
metastases. If LDH is to be used in the future as a
prognostic factor for survival for breast cancer patients Authors' Contributions
with bone metastases, it is desirable to agree on a com- Conception and design: J.E. Brown, A. Lipton, R.E. Coleman
Development of methodology: J.E. Brown, R.J. Cook, A. Lipton
mon way of expressing LDH values across centers. This Analysis and interpretation of data (e.g., statistical analysis, biosta-
study had a limited follow-up period and duration of tistics, computational analysis): J.E. Brown, R.J. Cook, A. Lipton
bisphosphonate use, being confined to the duration of Writing, review, and/or revision of the manuscript: J.E. Brown, R.J. Cook,
A. Lipton, R.E. Coleman
the original phase III study. The LDH assessment was Study supervision: R.E. Coleman
recorded at study entry and not at the time of diagnosis of
bone metastases (median difference 3.75 months). How- Acknowledgments
ever, given that the average survival for patients with The authors thank Cancer Research UK for support for J.E. Brown. The
advanced breast cancer was typically 18 to 26 months authors thank Michael Hobert, ProEd Communications, Inc., for medical
editorial assistance with this article.
after diagnosis of bone metastases, depending on hor-
monal therapy (23), the limited follow-up is not likely to
have any substantial impact on the survival data. Grant Support
Financial support for medical editorial assistance was provided by Novar-
On the basis of the current study, the combination of tis Pharmaceuticals Corporation, East Hanover, New Jersey.
LDH with other prognostic factors into a weighted prog- The costs of publication of this article were defrayed in part by the
nostic score for survival in patients with bone metastases payment of page charges. This article must therefore be hereby marked
advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate
from breast cancer would be a valuable next step. Such a this fact.
score would then require confirmation and validation in a
future prospective study before adoption into new man- Received May 7, 2012; revised August 21, 2012; accepted August 21, 2012;
agement guidelines. Assessment of LDH should, therefore, published OnlineFirst September 4, 2012.
References
1. Coleman RE. Metastatic bone disease: clinical features, pathophys- dures/DeviceApprovalsandClearances/Recently-ApprovedDevices/
iology and treatment strategies. Cancer Treat Rev 2001;27:165–76. ucm081239.htm. 2009.
2. Kinnane N. Burden of bone disease. Eur J Oncol Nurs 2007;11 suppl 2: 8. Stopeck AT, Lipton A, Body JJ, Steger GG, Tonkin K, de Boer RH, et al.
S28–31. Denosumab compared with zoledronic acid for the treatment of bone
3. Weinfurt KP, Castel LD, Li Y, Timbie JW, Glendenning GA, Schulman metastases in patients with advanced breast cancer: a randomized,
KA. Health-related quality of life among patients with breast cancer double-blind study. J Clin Oncol 2010;28:5132–9.
receiving zoledronic acid or pamidronate disodium for metastatic bone 9. Diamond JR, Finlayson CA, Borges VF. Hepatic complications of
lesions. Med Care 2004;42:164–75. breast cancer. Lancet Oncol 2009;10:615–21.
4. Sutherland CM, Mather FJ. Charity Hospital experience with 10. Tampellini M, Berruti A, Gerbino A, Buniva T, Torta M, Gorzegno G,
long-term survival and prognostic factors in patients with breast et al. Relationship between CA 15–3 serum levels and disease extent in
cancer with localized or regional disease. Ann Surg 1988;207: predicting overall survival of breast cancer patients with newly diag-
569–80. nosed metastatic disease. Br J Cancer 1997;75:698–702.
5. Dawood S, Broglio K, Buzdar AU, Hortobagyi GN, Giordano SH. 11. Dunphy FR, Spitzer G, Fornoff JE, Yau JC, Huan SD, Dicke KA, et al.
Prognosis of women with metastatic breast cancer by HER2 status Factors predicting long-term survival for metastatic breast cancer
and trastuzumab treatment: an institutional-based review. J Clin Oncol patients treated with high-dose chemotherapy and bone marrow
2010;28:92–8. support. Cancer 1994;73:2157–67.
6. Fan C, Oh DS, Wessels L, Weigelt B, Nuyten DS, Nobel AB, et al. 12. Kuru B, Camlibel M, Dinc S, Gulcelik MA, Gonullu D, Alagol H.
Concordance among gene-expression-based predictors for breast Prognostic factors for survival in breast cancer patients who devel-
cancer. N Engl J Med 2006;355:560–9. oped distant metastasis subsequent to definitive surgery. Singapore
7. U.S. Food and Drug Administration. CellSearch Epithelial Cell Kit/ Med J 2008;49:904–11.
CellSpotter Analyser - K031588 [cited 2012 16 April]. Available from: 13. Yavas O, Hayran M, Ozisik Y. Factors affecting survival in breast
http://www.fda.gov/MedicalDevices/ProductsandMedicalProce- cancer patients following bone metastasis. Tumori 2007;93:580–6.
14. Coleman RE, Smith P, Rubens RD. Clinical course and prognostic static breast cancer: a multivariate analysis [poster]. Presented at:
factors following bone recurrence from breast cancer. Br J Cancer EBCC 6: European Breast Cancer Conference; 2008 Apr 15–19; Berlin,
1998;77:336–40. Germany. Abstract nr 468.
15. Brown JE, Thomson CS, Ellis SP, Gutcher SA, Purohit OP, Coleman 31. Coleman RE, Rubens RD. The clinical course of bone metastases from
RE. Bone resorption predicts for skeletal complications in metastatic breast cancer. Br J Cancer 1987;55:61–6.
bone disease. Br J Cancer 2003;89:2031–7. 32. Coleman RE, Major P, Lipton A, Brown JE, Lee KA, Smith M, et al.
16. Brown JE, Cook RJ, Major P, Lipton A, Saad F, Smith M, et al. Bone Predictive value of bone resorption and formation markers in cancer
turnover markers as predictors of skeletal complications in prostate patients with bone metastases receiving the bisphosphonate zole-
cancer, lung cancer, and other solid tumors. J Natl Cancer Inst 2005; dronic acid. J Clin Oncol 2005;23:4925–35.
97:59–69. 33. Bacci G, Ferrari S, Longhi A, Rimondini S, Versari M, Zanone A, et al.
17. Lipton A, Cook R, Saad F, Major P, Garnero P, Terpos E, et al. Prognostic significance of serum LDH in Ewing's sarcoma of bone.
Normalization of bone markers is associated with improved survival Oncol Rep 1999;6:807–11.
in patients with bone metastases from solid tumors and elevated bone 34. Stokkel MP, van Eck-Smit BL, Zwinderman AH, Willems LN, Pauwels
resorption receiving zoledronic acid. Cancer 2008;113:193–201. EK. Pretreatment serum lactate dehydrogenase as additional staging
18. Lipton A, Demers L, Curley E, Chinchilli V, Gaydos L, Hortobagyi G, parameter in patients with small-cell lung carcinoma. J Cancer Res Clin
et al. Markers of bone resorption in patients treated with pamidronate. Oncol 1998;124:215–9.
Eur J Cancer 1998;34:2021–6. 35. Cerny T, Blair V, Anderson H, Bramwell V, Thatcher N. Pretreatment
19. Major PP, Cook RJ, Lipton A, Smith MR, Terpos E, Coleman RE. prognostic factors and scoring system in 407 small-cell lung cancer
Natural history of malignant bone disease in breast cancer and the use patients. Int J Cancer 1987;39:146–9.
of cumulative mean functions to measure skeletal morbidity. BMC 36. Eigentler TK, Figl A, Krex D, Mohr P, Mauch C, Rass K, et al. Number of
Cancer. 2009;9:272. metastases, serum lactate dehydrogenase level, and type of treatment
20. Kaminski M, Rosen LS, Gordon D, Zheng M, Hei Y-J. Zoledronic acid is are prognostic factors in patients with brain metastases of malignant
superior to pamidronate in patients with breast cancer and multiple melanoma. Cancer 2011;117:1697–703.
myeloma who are at high risk for skeletal complications [poster]. 37. Egberts F, Kotthoff EM, Gerdes S, Egberts JH, Weichenthal M, Haus-
Presented at: Primary Therapy of Early Breast Cancer 9th International child A. Comparative study of YKL-40, S-100B and LDH as monitoring
Conference; 2005 Jan 26–29; St. Gallen, Switzerland. Abstract 107. tools for Stage IV melanoma. Eur J Cancer 2012;48:695–702.
21. Rosen LS, Gordon D, Kaminski M, Howell A, Belch A, Mackey J, et al. 38. Chibaudel B, Bonnetain F, Shi Q, Buyse M, Tournigand C, Sargent DJ,
Zoledronic acid versus pamidronate in the treatment of skeletal metas- et al. Alternative end points to evaluate a therapeutic strategy in
tases in patients with breast cancer or osteolytic lesions of multiple advanced colorectal cancer: evaluation of progression-free survival,
myeloma: a phase III, double-blind, comparative trial. Cancer J duration of disease control, and time to failure of strategy—an Aide et
2001;7:377–87. Recherche en Cancerologie Digestive Group Study. J Clin Oncol
22. Rosen LS, Gordon DH, Dugan W Jr, Major P, Eisenberg PD, Pro- 2011;29:4199–204.
vencher L, et al. Zoledronic acid is superior to pamidronate for the 39. Terpos E, Katodritou E, Roussou M, Pouli A, Michalis E, Delimpasi S,
treatment of bone metastases in breast carcinoma patients with at et al. High serum lactate dehydrogenase adds prognostic value to the
least one osteolytic lesion. Cancer 2004;100:36–43. international myeloma staging system even in the era of novel agents.
23. Rosen LS, Gordon D, Kaminski M, Howell A, Belch A, Mackey J, et al. Eur J Haematol 2010;85:114–9.
Long-term efficacy and safety of zoledronic acid compared with 40. Yamamoto N, Watanabe T, Katsumata N, Omuro Y, Ando M,
pamidronate disodium in the treatment of skeletal complications in Fukuda H, et al. Construction and validation of a practical prog-
patients with advanced multiple myeloma or breast carcinoma: a nostic index for patients with metastatic breast cancer. J Clin Oncol
randomized, double-blind, multicenter, comparative trial. Cancer 1998;16:2401–8.
2003;98:1735–44. 41. Le Scodan R, Massard C, Jouanneau L, Coussy F, Gutierrez M, Kirova
24. Brown JE, Cook RJ, Lipton A, Costa L, Coleman RE. Prognostic factors Y, et al. Brain metastases from breast cancer: proposition of new
for skeletal complications from metastatic bone disease in breast prognostic score including molecular subtypes and treatment. J Neu-
cancer. Breast Cancer Res Treat 2010;123:767–79. rooncol 2012;106:169–76.
25. Albain KS, Crowley JJ, LeBlanc M, Livingston RB. Survival determi- 42. Tredan O, Ray-Coquard I, Chvetzoff G, Rebattu P, Bajard A, Chabaud
nants in extensive-stage non-small-cell lung cancer: the Southwest S, et al. Validation of prognostic scores for survival in cancer patients
Oncology Group experience. J Clin Oncol 1991;9:1618–26. beyond first-line therapy. BMC Cancer 2011;11:95.
26. Eleutherakis-Papaiakovou V, Bamias A, Gika D, Simeonidis A, Pouli 43. Cook RJ, Coleman R, Brown J, Lipton A, Major P, Hei YJ, et al. Markers
A, Anagnostopoulos A, et al. Renal failure in multiple myeloma: of bone metabolism and survival in men with hormone-refractory
incidence, correlations, and prognostic significance. Leuk Lympho- metastatic prostate cancer. Clin Cancer Res 2006;12:3361–7.
ma 2007;48:337–41. 44. Naruse K, Yamada Y, Aoki S, Taki T, Nakamura K, Tobiume M,
27. Hoster E, Dreyling M, Klapper W, Gisselbrecht C, van Hoof A, Kluin- et al. Lactate dehydrogenase is a prognostic indicator for prostate
Nelemans HC, et al. A new prognostic index (MIPI) for patients with cancer patients with bone metastasis. Hinyokika Kiyo 2007;53:
advanced-stage mantle cell lymphoma. Blood 2008;111:558–65. 287–92.
28. Smaletz O, Scher HI, Small EJ, Verbel DA, McMillan A, Regan K, et al. 45. Yamada Y, Nakamura K, Aoki S, Tobiume M, Zennami K, Kato Y, et al.
Nomogram for overall survival of patients with progressive meta- Lactate dehydrogenase, Gleason score and HER-2 overexpression
static prostate cancer after castration. J Clin Oncol 2002;20:3972–82. are significant prognostic factors for M1b prostate cancer. Oncol Rep
29. Tas F, Aydiner A, Demir C, Topuz E. Serum lactate dehydrogenase 2011;25:937–44.
levels at presentation predict outcome of patients with limited-stage 46. Scher HI, Jia X, de Bono JS, Fleisher M, Pienta KJ, Raghavan D, et al.
small-cell lung cancer. Am J Clin Oncol 2001;24:376–8. Circulating tumour cells as prognostic markers in progressive, cas-
30. Coleman RE, Brown JE, Cook R. Serum lactate dehydrogenase (LDH) tration-resistant prostate cancer: a reanalysis of IMMC38 trial data.
is a significant prognostic variable for survival in patients with meta- Lancet Oncol 2009;10:233–9.
Serum Lactate Dehydrogenase Is Prognostic for Survival in

Patients with Bone Metastases from Breast Cancer: A
Retrospective Analysis in Bisphosphonate-Treated Patients
Janet E. Brown, Richard J. Cook, Allan Lipton, et al.
Clin Cancer Res 2012;18:6348-6355. Published OnlineFirst September 4, 2012.
Updated version Access the most recent version of this article at:
doi:10.1158/1078-0432.CCR-12-1397
Cited articles This article cites 43 articles, 9 of which you can access for free at:
http://clincancerres.aacrjournals.org/content/18/22/6348.full#ref-list-1
E-mail alerts Sign up to receive free email-alerts related to this article or journal.
Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at
Subscriptions [email protected].
Permissions To request permission to re-use all or part of this article, use this link
http://clincancerres.aacrjournals.org/content/18/22/6348.
Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
Rightslink site.

6348 Full

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

6348 Full

Uploaded by

Copyright:

Available Formats

Published OnlineFirst September 4, 2012; DOI: 10.1158/1078-0432.

Serum Lactate Dehydrogenase Is Prognostic for Survival in

Introduction cemia of malignancy. Overall, SREs have been associated with

6348 Clin Cancer Res; 18(22) November 15, 2012

Prognostic Value of LDH in Patients with Bone Metastases

III trial comparing ZOL with pamidronate (21–23). All

www.aacrjournals.org Clin Cancer Res; 18(22) November 15, 2012 6349

into <60, 60 to <70, 70 to <80, and 80 kilograms.

Prognostic Value of LDH in Patients with Bone Metastases

Univariate Full multivariate

Factor RR 95% CI P RR 95% CI P

Abbreviations: HT, hormone therapy; RR, risk ratio.

www.aacrjournals.org Clin Cancer Res; 18(22) November 15, 2012 6351

Prognostic Value of LDH in Patients with Bone Metastases

patients (n ¼ 299) with metastatic cancer treated at a single

www.aacrjournals.org Clin Cancer Res; 18(22) November 15, 2012 6353

Prognostic Value of LDH in Patients with Bone Metastases

www.aacrjournals.org Clin Cancer Res; 18(22) November 15, 2012 6355

Serum Lactate Dehydrogenase Is Prognostic for Survival in

Clin Cancer Res 2012;18:6348-6355. Published OnlineFirst September 4, 2012.

You might also like