SEARCH

SEARCH BY CITATION

Keywords:

  • DEAD box 1 (DDX1);
  • radiation therapy;
  • molecular markers;
  • ipsilateral breast tumor recurrence (IBTR);
  • early stage breast cancer

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conclusions
  7. FUNDING SOURCES
  8. REFERENCES

BACKGROUND:

DEAD box 1 (DDX1) is an RNA helicase with a number of roles, including translation initiation, RNA splicing and modification, and possibly DNA double-strand break repair. Amplification of DDX1 expression has been implicated in tumors including neuroblastoma, Wilms tumor, retinoblastoma, and testicular carcinoma. The purpose of this study was to evaluate the prognostic significance of DDX1 expression in patients with breast cancer treated with breast-conserving therapy.

METHODS:

Paraffin-embedded specimens from 282 women with node-negative stage 1 and 2 breast cancer treated with breast-conserving surgery and radiation therapy were constructed into tissue microarrays and stained for DDX1. The molecular profiles were correlated with clinicopathologic factors and overall, local, and distant metastatic-free survival.

RESULTS:

DDX1 positivity was identified in 142 (50%) patients. The median age at diagnosis was 53 years. Eighty percent of the patients had T1 disease; 11% were HER2neu-positive, and 18% had triple-negative disease. DDX1 negativity was strongly associated with triple-negative phenotype (P = .01). DDX1 positivity was found to be associated with improved local relapse-free survival (96% vs 85%, P = .0233), distant metastatic-free survival (95% vs 85%, P = .0320), and overall survival (92% vs 84%, P = .0474) at 10 years.

CONCLUSIONS:

Node-negative, early stage breast cancer patients with high levels of DDX1 were found to have a significant improvement in local control, distant metastatic-free survival, and overall survival compared with patients with low levels of DDX1. Cancer 2012;. © 2011 American Cancer Society.

The American Cancer Society estimated that nearly 207,000 cases of invasive breast cancer would be diagnosed in 2010.1 Breast-conserving surgery (BCS) followed by radiation therapy (RT) has the become standard of care for early stage breast cancer, with outcomes similar to mastectomy in many randomized clinical trials.2 Breast tumors can be individually characterized by many different factors including tumor stage, nodal status, estrogen receptor (ER)/progesterone receptor (PR) status, HER2 status, tumor grade, proliferation with Ki-67, and molecular subtype, each of which may have prognostic significance and may help create tailored treatment regimens for each patient. The identification of prognostic molecular markers has allowed the use of anti-ER therapeutic agents such as tamoxifen and anastrazole, in addition to highly specific monoclonal antibody therapeutic agents such as trastuzumab to improve outcomes of breast cancer.3, 4 Oncotype DX is a 21-gene assay with both prognostic and predictive value in women with early stage ER-positive, node-negative disease. The assay assigns a recurrence score that allows clinicians to stratify patients into separate risk categories and determine who may benefit the most from adjuvant chemotherapy.5 The use of molecular markers in the same manner as prognostic tools for radiotherapy is being investigated.6

Another potential marker is DEAD box 1 (DDX1). DDX1, also known as DEAD-box protein-retinoblastoma, is in the family of DEAD-box proteins, a group of adenosine triphosphate–dependent RNA helicases named for the conserved D-E-A-D (Asp-Glu-Ala-Asp) amino acid sequence.7 DEAD-box proteins are highly conserved in many organisms ranging from Escherichia coli to higher mammals, and DDX1 messenger RNA (mRNA) is expressed in many human tissues.8 DEAD-box family proteins have a wide role in normal cellular function including translation initiation, mRNA synthesis, RNA splicing and modification, and assembly of ribosomes and spliceosomes.9 In cell lines exposed to ionizing radiation, DDX1 rapidly redistributes to areas of double-strand breaks and DNA damage in RNA-DNA structures, possibly assisting in the repair of transcriptionally active DNA.10 Knockout of DDX1 in Drosophila melanogaster results in a lethal phenotype.11

DDX1 has already been implicated in a number of cancer types and metastases. DDX1 is known to associate with heterogeneous nuclear ribonucleoprotein K, the accumulation of which is important in cancer metastasis.12 Coamplification of DDX1 and the MYC proto-oncogene has been characterized in retinoblastoma, neuroblastoma, Wilms tumor, and alveolar rhabdomyosarcoma cell lines and tissue.13-15 DDX1 mRNA was amplified in both seminomatous and nonseminomatous samples of human testicular germ cell tumors. In nude nice, testicular germ cell tumors with knocked-down DDX1 were unable to form solid tumors, suggesting that DDX1 is required for tumorigenesis.8 In human breast cancer, elevated expression of DDX1 mRNA and cytoplasmic DDX1 levels were associated with breast cancer relapse, and high expression was associated with poor survival.16, 17

Li et al10 found that after exposure ionizing radiation, DDX1 in cell lines redistributes to areas of double-stranded breaks and DNA damage. Because DDX1 is recruited within minutes of exposure of ionizing radiation, it is thought to play a role in DNA double-strand break repair. DDX1 may be a marker of radiotherapy resistance, because double-strand break induction is a primary method by which radiotherapy promotes cancer cell death. Previous studies have largely reported increased DDX1 mRNA expression in patients with testicular cancer, neuroblastoma, and retinoblastoma. There is only 1 study to date that has addressed DDX1 expression in tissue microarrays with treatment-naive breast cancer samples.16

To our knowledge, there has not been an analysis of DDX1 expression as a prognostic factor in women with early stage breast cancer treated with BCS and adjuvant RT. The purpose of the present study was to analyze the prognostic significance of DDX1 in a large cohort of node-negative, early stage breast cancer patients treated with BCS and RT.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conclusions
  7. FUNDING SOURCES
  8. REFERENCES

Patient Selection

Patients were treated at the Yale University Department of Therapeutic Radiology between 1975 and 2003. A total of 282 patients met the following inclusion criteria: (1) node-negative stage 1 or 2 breast cancer treated with BCS followed by RT to the intact breast and (2) primary breast cancer tissue available for study in paraffin-embedded blocks from the archives of the hospital or referring hospitals for processing into a tissue microarray. Information about each patient's clinical history was obtained from patient charts and assembled into a database. The size of the primary tumor was defined as the largest tumor diameter reported by the pathologist after surgery. Lymph node status was determined by histological evidence of lymph node metastases. The primary endpoints of the study were local recurrence, distant metastasis-free survival, and overall survival. Local recurrence was defined as clinical or biopsy-proven tumor recurrence in the ipsilateral breast. The study was conducted after obtaining approval from the Yale Human Investigations Committee.

All patients in this study were treated with BCS with or without axillary lymph node dissection, as clinically indicated and based on the standard practice patterns of treating surgeons throughout the interval. After BCS, all patients were treated with daily fractionated RT to the intact whole breast. A third field to treat the regional nodes to a dose of 46 Gy was included as clinically indicated at the discretion of the treating radiation oncologist. Patients were treated using 4- to 6-MeV photons with a daily fraction size of 2.0 Gy to a total median dose of 48.0 Gy to the whole breast, routinely followed by an electron cone-down to the lumpectomy cavity to a total median dose of 64.0Gy. Adjuvant chemotherapy was administered to patients as clinically indicated in accordance to the standard practice of medical oncologists during this interval. Adjuvant hormone therapy was routinely given to ER-positive patients.

Construction of Tissue Microarray

All 282 patients in our study had tissue available in a tissue microarray constructed for molecular marker expression in breast cancer tissue. A pathologist examined hematoxylin- and eosin-stained slides of the archived paraffin blocks of breast cancer tissue and circled representative tumor sections. Areas of tumor that were distinct from normal epithelium were identified and marked for subsequent analysis. From these tumor sections, two 0.6-mm cores were extracted using a tissue microarray device (Beecher Instruments, Silver Spring, MD). There were a total of 484 cores for the 282 patients in the study. Microarrays were cut into 5-μm-thick sections with a tape-based tissue transfer system (Intrumedics, Hackensack, NJ) and processed onto slides.

Staining

Immunohistochemical analysis was performed on 5-μm thick tissue sections prepared from formalin-fixed, paraffin-embedded archival tissue from the tissue microarray block constructed. Tissue sections were deparaffinized and then quenched in 2% hydrogen peroxide–methanol solution. Samples were then pretreated to promote antigen retrieval with the DAKO Target Retrieval Solution (DAKO, Carpinteria, CA). A 3% hydrogen peroxide solution was then used for endogenous peroxidase blocking. Slides were then incubated with rabbit polyclonal antibody DDX1 (1:250; Bethyl Laboratories, Montgomery, TX). Slides were additionally incubated overnight at 4°C with the following antibodies: 1) ER, mouse monoclonal antihuman ER (DAKO); 2) PR, mouse monoclonal antihuman PR (DAKO); 3) HER2neu: rabbit polyclonal anti-HER-2/neu oncoprotein (DAKO). Slides were then incubated with secondary antibody, labeled with avidin-biotin complex streptavidin-peroxidase (Elite; Vector Laboratories, Burlingame, CA), and incubated with the chromogen diaminobenzidine tetrahydrochloride as a chromogenic substrate. Finally, slides were counterstained with hematoxylin, dehydrated with ethanol, and mounted. A known positive case was included as a positive control. For the negative control, the primary antibody was replaced with nonimmune mouse serum.

Quantitative and qualitative assessment of all 4 biomarkers stained was performed by a single experienced pathologist (H.W.) who was blinded to patient outcomes. For cores that were uninterpretable because of tissue loss or lack of tumor cells, a score of “not applicable” was given. For each core, the region of predominant staining intensity was scored. ER and PR were assessed by the number of positive-stained nuclei. For HER2neu, only membrane staining was scored positive. A numeric score ranging from 0 to 3 that reflected the staining intensity and patterns in 10% or more tumor was used. For HER2 status, numeric scores of 2 or 3 were considered positive. The DDX1 staining was primarily cytoplasmic, with slight nuclear involvement. The intensity of DDX1 staining was scored as 0 (no immunoreactivity), 1 (weak), 2 (moderate), or 3 (strong) in tumor cells. Cases scored at 0 and 1 were considered as a group as negative expression levels, whereas cases scored at 2 and 3 were considered positive. This cutoff was chosen because it is routinely used in our laboratory and was used in a previously published report on DDX1 expression in breast cancer tissue.16

Statistical Analysis

The independent variables for analysis included age, race, tumor stage, nodal status, triple-negative status (defined as ER-negative, PR-negative, and HER2-negative), ER and PR positivity, surgical margin status, treatment with adjuvant chemotherapy and hormone therapy, and DDX1 expression. The molecular profiles of these patients were correlated with clinicopathologic factors, local relapse-free survival (LRFS), distant metastasis-free survival (DMFS), disease-free survival (DFS), cause-specific survival (CSS), and overall survival (OS). Median follow-up was calculated using the Kaplan-Meier product-limit method, and differences in survival were ascertained using the log-rank method. Bivariate analyses for the association between covariables and DDX1 status included the chi-square test and Fisher exact test. The association between DDX1, covariables, and outcome was assessed in a multivariate model using the Cox proportional hazard regression model. Data on DDX1 status and relevant covariables were assembled in a database and analyzed using SAS version 9.2 software (SAS Institute, Cary, NC). All tests of statistical significance were 2-sided, and P<.05 was considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conclusions
  7. FUNDING SOURCES
  8. REFERENCES

Patient Characteristics

A total of 282 women with node-negative, early stage breast cancer treated with BCS were included in the study. The median age at diagnosis was 53 years (range, 28-87 years), with 42% of the patients under the age of 50 years at the time of diagnosis. Eighty percent of the patients had stage T1 disease; 46% and 35% received adjuvant hormonal therapy and adjuvant chemotherapy, respectively; 61% and 48% had ER-positive and PR-positive disease, respectively; 11% were HER2neu-positive; and 18% had triple-negative disease. As of September 2008, there was a median follow-up time of 7.3 years (range, 1.0-25.9 years). The ipsilateral breast relapse-free rate for the whole cohort at 5 and 10 years was 94% and 90%, with DMFS of 94% and 89% and OS of 94% and 88%, respectively. Additional clinical and pathologic information is provided in Table 1.

Table 1. Patient Clinical and Pathologic Characteristics
CharacteristicsNo. (%)
  1. ER, estrogen receptor; PR, progesterone receptor.

Age, y 
 ≤50118 (42)
 >50164 (58)
Race 
 White236 (84)
 Black36 (13)
 Other10 (3)
Smoking history 
 Negative140 (50)
 Positive72 (25)
 Unknown70 (25)
Tumor stage 
 T1226 (80)
 T253 (19)
 Other/Unknown3 (1)
Surgical margin status 
 Negative240 (85)
 Close/Positive18 (6)
 Unknown24 (9)
Received adjuvant hormone therapy 
 No151 (54)
 Yes128 (46)
Received adjuvant chemotherapy 
 No183 (65)
 Yes97 (35)
Estrogen receptor status 
 Negative99 (35)
 Positive171 (61)
 Unknown12 (4)
Progesterone receptor status 
 Negative105 (37)
 Positive135 (48)
 Unknown42 (15)
HER2 status 
 Negative218 (77)
 Positive31 (11)
 Unknown33 (12)
Triple-negative status 
 No201 (71)
 Yes51 (18)
 Unknown30 (11)
Molecular subtype 
 ER/PR+, HER2−135 (48)
 ER/PR+, HER2+16 (6)
 ER/PR−, HER2−51 (18)
 ER/PR−, HER2+13 (5)
 Unknown67 (24)

Immunohistochemical Staining Results and Correlation With Clinicopathologic Variables

Intracellular DDX1 immunohistochemical staining of tumor cells showed predominantly cytoplasmic localization, but there was some nuclear reactivity as well. The intracellular distribution of staining for DDX1 was heterogeneous. There was little or no immunoreactivity in some tumor specimens, but there was extensive staining of tumor cells in other specimens. Using the criteria for positivity as discussed previously, DDX1 expression was positive in 142 (50.3%) of the 282 patients. A total of 22 samples were uninterpretable due to missing tissue and were subsequently eliminated from survival analysis. Representative immunostaining results are shown in Figure 1.

thumbnail image

Figure 1. Representative images of (A) DEAD box 1–positive tumor and (B) DEAD box 1–negative tumor specimens are shown.

Download figure to PowerPoint

Using the chi-square and Fisher exact tests, DDX1-positive status was significantly associated with PR positivity (P = .03), and DDX1-negative status was significantly associated with triple-negative phenotype (P = .01). DDX1 status was not significantly correlated and did not approach significance with other variables, including age >50 years, race, smoking status, tumor stage, surgical margin status, ER status, HER2neu status, and reception of either adjuvant chemotherapy or hormonal therapy (Table 2).

Table 2. Chi-Square Analysis
CharacteristicsDDX1-Negative (%)DDX1-Positive (%)P
  1. DDX1, DEAD box 1; ER, estrogen receptor; OS, overall survival; PR, progesterone receptor.

  2. Data are expressed as no. (%).

Age, y   
 ≤5056 (22)55 (21).1567
 >5062 (24)87 (33) 
Race   
 White103 (40)116 (45).2175
 Nonwhite15 (6)26 (10) 
Smoking   
 Negative104 (28)134 (36).1667
 Positive70 (19)67 (18) 
Tumor stage   
 T198 (38)114 (44).7794
 T219 (7)26 (10) 
Surgical margin status   
 Negative99 (42)121 (51).8695
 Close/Positive8 (3)9 (4) 
ER status   
 Negative45 (18)46 (18).2818
 Positive67 (27)91 (37) 
PR status   
 Negative52 (24)45 (20).0310
 Positive48 (22)75 (34) 
HER2 status   
 Negative97 (41)112 (47).1539
 Positive9 (4)19 (8) 
Triple-negative status   
 No77 (33)109 (46).0132
 Yes30 (13)19 (8) 
Received adjuvant hormone therapy   
 No64 (25)73 (28).6823
 Yes53 (21)67 (26) 
Received adjuvant chemotherapy   
 No78 (30)91 (35).7204
 Yes39 (15)50 (19) 
Molecular subtype   
 Luminal A (ER/PR+, HER2−)54 (21)75 (29).1137
 Luminal B (ER/PR+, HER2+)4 (2)10 (4) 
 Triple-negative (ER/PR−, HER2−)30 (12)19 (7) 
 HER2+ (ER/PR−, HER2+)5 (2)8 (3) 

Association Between Clinical and Pathologic Variables and Patient Outcomes

On univariate analysis, positive DDX1 expression predicted improved outcomes in LRFS (hazard ratio [HR], 0.347; P = .03), DMFS (HR, 0.367; P = .04), and OS (HR, 0.434; P = .054) (Table 3). Positive DDX1 expression approached significance for improved CSS (HR, 0.361; P = .055). Age >50 years was a predictor of improved LRFS (HR, 0.333; P = .0158) and DMFS (HR, 0.438; P = .0502) but also predicted a trend toward poorer OS (HR, 2.31; P = .0556). In addition, stage T2 tumors predicted poorer DMFS (HR, 2.237; P = .0104).

Table 3. Univariate Analysis of DDX1 Expression
CharacteristicLRFSDMFSCSSOS
 HRPHRPHRPHRP
  1. CSS, cause-specific survival; DDX1, DEAD box 1; DMFS, distant metastasis-free survival; ER, estrogen receptor; HR, hazard ratio; LRFS, local relapse-free survival; OS, overall survival; PR, progesterone receptor; Ref., reference group.

Age, y        
 ≤50 (Ref.)0.333.01580.438.05021.484.39432.31.0556
 >50        
Race        
 White (Ref.)0.989.9860.884.84140.32.26570.236.1558
 Black        
Smoking history        
 Negative (Ref.)1.602.36511.67.35723.041.03491.795.181
 Positive        
Tumor stage        
 T1 (Ref.)1.519.33992.237.01041.745.15451.773.0947
 T2        
Surgical margin status        
 Negative (Ref.)1.935.38330.655.68050.796.82420.524.5266
 Close/Positive        
Adjuvant hormone therapy        
 No (Ref.)0.58.23490.334.02920.411.0830.726.4183
 Yes        
Adjuvant chemotherapy        
 No (Ref.)1.005.99161.361.46640.544.27490.534.2072
 Yes        
ER status        
 Negative (Ref.)1.637.3040.617.23711.284.58991.344.4655
 Positive        
PR status        
 Negative (Ref.)1.182.69270.691.38061.028.95051.061.8761
 Positive        
HER2 status        
 Negative (Ref.)0.822.79742.086.19480.994.99330.693.621
 Positive        
Luminal A (ER/PR+, HER2−)        
 Negative (Ref.)0.836.68080.615.26251.404.60161.4868.2969
 Positive        
Luminal B (ER/PR+, HER2+)        
 Negative (Ref.)0.710.73820.98810.98880.9870
 Positive        
Triple-negative status        
 Negative (Ref.)1.243.64791.047.92750.642.48120.829.7063
 Positive        
HER2+ (ER/PR−, HER2+)        
 Negative (Ref.)0.841.87015.029.00322.393.24061.780.4324
 Positive        
DDX1 expression        
 Negative (Ref.)0.347.03020.367.03980.361.05540.434.054
 Positive        

Using log-rank survival analysis, positive DDX1 expression was again associated with improved LRFS, DMFS, CSS, and OS (Figure 2). Positive DDX1 expression showed improved survival at 10-year follow-up for all survival measures (Table 4).

thumbnail image

Figure 2. Kaplan-Meier survival analysis is shown for DEAD box 1–positive (red) versus DEAD box 1–negative (black) cohorts, respectively. (A) Breast relapse-free survival (P = .02). (B) Distant metastasis-free survival (P = .03). (C) Disease-free survival (P = .0077). (D) Cause-specific survival (P = .045). (E) Overall survival (P = .047).

Download figure to PowerPoint

Table 4. Patient Outcomes (% Surviving) at 10-Year Follow-up
 DDX1- PositiveDDX1- NegativePa
  • DDX1, DEAD box 1.

  • a

    Log-rank test.

Overall survival9284.0474
Local relapse-free survival9685.0233
Distant metastasis-free survival9585.0320
Disease-free survival9074.0077
Cause-specific survival9385.0454

DDX1 status, tumor stage, ER positivity, PR positivity, age >50 years, and HER2neu expression were entered into a multivariate model. Multivariate survival analysis was conducted using the Cox proportional hazards model and demonstrated that DDX1 expression approached significance (P = .1) as an independent prognostic marker for LRFS, DMFS, CSS, and OS (Table 5).

Table 5. Multivariate Analysis of DDX1 Expression
 LRFSDMFSDFSCSSOS
  • CSS, cause-specific survival; DDX1, DEAD box 1; DFS, disease-free survival; DMFS, distant metastasis-free survival; ER, estrogen receptor; LRFS, local relapse-free survival; OS, overall survival; PR, progesterone receptor.

  • All data are expressed as P values.

  • a

    Molecular subtype based on ER, PR, and HER2 status.

Tumor stage.8500.1084.3436.3344.1181
Age.0095.9056.0510.1950.0260
ER status.9921.3362.6678.9930.9917
PR status.6708.7428.9028.3430.2638
HER2.9922.2667.5931.9934.9913
Subtypea.9926.7166.4702.9928.9913
DDX1.2301.1326.1171.0914.0895

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conclusions
  7. FUNDING SOURCES
  8. REFERENCES

The majority of women diagnosed with breast cancer will present with early stage invasive disease and will require RT after BCS. Even with the addition of RT after BCS, there remains a 10%-20% risk of local recurrence.18 Breast cancer, even of the same cancer stage and primary histology, is a clinically heterogeneous disease. The heterogeneity may be attributed to differences in molecular profile of the tumor, which may result in differences in prognosis, as well as response to treatment. The major molecular subtypes of breast cancer are already known to have differences in long-term survival.19, 20 The use of molecular markers as prognostic indicators has been studied increasingly to improve prediction of clinical outcomes after BCS and RT.

DDX1 is a highly conserved RNA helicase involved in many normal cellular functions, including DNA double-strand break repair. DDX1 expression has already been associated with various cancers, including retinoblastoma, neuroblastoma, and testicular carcinoma. This is the first study of DDX1 expression in a large cohort of women with early stage breast cancer treated only with breast-conserving therapy (BCT). In this study, we analyzed expression of DDX1 in a cohort of patients treated with BCT after diagnosis of early stage breast cancer. We found that approximately 50% of patients were positive for DDX1 expression using our aforementioned criteria; similarly, Germain et al16 found approximately 33% positivity using comparable scoring criteria. DDX1-negative status was also significantly associated with triple-negative status, an aggressive molecular subtype of breast cancer. In both univariate and log-rank analysis of Kaplan-Meier survival curves, DDX1 positivity significantly predicted improved LRFS, DMFS, and OS.

With the exception of triple-negative status and PR status, we were unable to detect any other clinical or pathologic measures that correlated with DDX1 positivity in our cohort, including age, race, tumor stage, ER status, HER2neu status, treatment with adjuvant chemotherapy, or treatment with adjuvant hormonal therapy. DDX1 positivity lost significance in our multivariate model.

Although we found DDX1 positivity to have a significant effect on improved LRFS, DMFS, and OS, further study of DDX1 is needed to elucidate a mechanism for heterogenous expression of DDX1 and improved outcomes in patients treated with BCT. Li et al have shown that exposure to ionizing radiation causes rapid formation of numerous DDX1 foci that colocalize with γ-H2AX and phosphorylated ataxia telangiectasia mutated (ATM) foci at DSB sites.10 The formation of DDX1 foci at site of DNA damage after ionizing radiation exposure is dependent on ATM, which is known to activate a number of proteins in the DNA damage checkpoint pathway, including tumor suppressors p53 and CHK2; DDX1 may represent another important member of this tumor suppression pathway.21 It has been proposed that DDX1 acting as an RNAse and RNA-DNA unwinder may facilitate the clearance of RNA from areas of DNA double-strand breaks (DSBs) due to ionizing radiation or radiomimetic drugs such as bleomycin, thus allowing for DNA DSB repair.10 Even though we know that DDX1 is important in DNA and RNA metabolism and may be implicated in DSB repair, the role of cytoplasmic DDX1, especially its molecular targets, remains unknown and represents an exciting area of study in DNA repair.

Previous reports have found elevated levels of DEAD-box proteins in many tumors including breast cancer, testicular carcinoma, and neuroblastoma.8 In neuroblastoma, DDX1 is often, but not always, coamplified with the MYCN oncogene, due to close physical proximity to the MYCN oncogene.22, 23 MYCN amplification in neuroblastoma is strongly associated with poor disease prognosis. However, large series have shown that DDX1 coamplification with MYCN can actually result in improved clinical outcome. This finding suggests that DDX1 amplification rescues patients with tumors that amplify MYCN, and even that DDX1 expression may make tumors in general more susceptible to therapy.24-26 Other DEAD-box proteins have previously been found as transcriptional coactivators of ERα; furthermore, DDX17 plays a critical role in ERα coactivation and subsequent estrogen-dependent tumorigenesis. Tumors that express ER have already been associated with improved clinical outcomes.27 Although our study did not find a similar association between DDX1 expression and ER expression, we did find that DDX1-negative tumors were more likely to be triple-negative tumors compared with DDX1-positive tumors. Our report found that DDX1 expression improved clinical outcomes, and future studies must confirm that effect is solely due to DDX1 or interaction with other pathways.

Germain et al16 conducted the only other analysis to date of DDX1 in treatment-naive breast cancer tissue microarrays. In their case-control study, 88 patients who experienced early relapse (defined as recurrence within 5 years) were matched to 88 patients who did not experience early relapse according to ER status, HER2neu status, stage, and follow-up. The study included 45 stage 1 patients, 117 stage 2A/2B patients, and 14 stage 3A/3B patients. Fifty-six of the 176 tumors (32%) were classified as triple-negative, and 113 of the 176 patients (64%) had breast cancer tissue available for immunohistochemistry. Using mRNA expression and DDX1 cytoplasmic positivity, elevated DDX1 mRNA expression and DDX1 protein in the cytoplasm was significantly associated with negative ER status and negative PR status and also predicted early breast cancer recurrence in univariate, multivariate, and log-rank analysis.16

Our study differs from the Germain et al16 study in several important ways. First, the present study was a cohort study of node negative, early stage breast cancer consisting of a relatively homogenous population of 81% stage 1 and 19% stage 2A patients. Germain et al performed a matched pair analysis in which they selected a cohort of 88 patients who experienced an early relapse (defined as relapse at <5 years after treatment) out of a cohort of 988 patients. An additional 88 patients who did not have early relapse were matched on ER and HER2 status, stage, and time of follow-up and were included for analysis. Overall, their population consisted of 26% stage 1, 66% stage 2A/2B, and 8% stage 3A/3B patients. This is important because advanced tumor stage and nodal stage are independently associated with poorer overall survival in breast cancer.21 Furthermore, our study analyzed cytoplasmic DDX1 protein expression using immunohistochemistry, whereas Germain et al investigated DDX1 RNA expression in 176 patients, and only quantitated cytoplasmic protein expression in 113 of 176 patients. Thus, our study had more than twice as many tissue samples available for immunohistochemical analysis. Finally, the median follow-up in our cohort was approximately 3 years longer. Differences in the results of our studies may be due to the inherent differences in study populations and their cohort having more advanced disease.

In addition, the differences in results may be due to altered expression of DDX1 as breast cancer progresses. The previous study had a higher proportion of advanced disease and disease with indicators of poor prognosis, such as triple-negative phenotype. It is possible that DDX1 expression may confer improved prognosis during early disease as in our study, but poorer prognosis in advanced disease as in the previous study. Germain et al have suggested that deregulation of DDX1 is the key step in its association with recurrence, but the role of cytoplasmic DDX1 expression is not yet well understood and must be studied further. Any attempt to use cytoplasmic DDX1 expression prospectively as a prognostic marker should be preceded by rigorous analysis of DDX1 expression in both patients with early stage and locally advanced disease.13, 17

This present study is the largest tissue microarray analysis of DDX1 expression in breast cancer, a protein that plays a critical role in normal cellular metabolism and may play an important role in DNA DSB repair. Compared with previous reports, this study is strengthened by its relatively homogeneous population: all women were diagnosed with stage 1 and 2 node-negative breast cancer and were treated with BCS and RT at a single institution. Our report is further strengthened because the median follow-up time was approximately 3 years longer than the previous study of DDX1 in breast cancer. Our study was retrospective in nature, and the results should be considered a rigorous early analysis of the prognostic significance of DDX1 in this cohort. Confirmation of the results are warranted in additional tissue microarray analyses of similar cohorts of breast cancer patients treated with BCT to corroborate DDX1 positivity as an indicator of good prognosis in early stage breast cancer survivors. Rigorous prospective clinical trials may be used both to confirm our results as well as study whether DDX1-negative patients, particularly those who have triple-negative disease, may benefit from more aggressive adjuvant chemotherapy or RT.

Conclusions

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conclusions
  7. FUNDING SOURCES
  8. REFERENCES

Our results suggest that in patients with node-negative, early stage breast cancer treated with BCT, DDX1 positivity predicts significant improvement in LRFS, DMFS, and OS. In addition, our study indicates that DDX1 negativity is strongly associated with the aggressive triple-negative molecular subtype of breast cancer, consistent with patients with DDX1-negative tumors having poorer local control and overall survival. Further study is required to understand the mechanism behind DDX1 positivity and improved survival, and what targeted therapies, if any, may be used to improve clinical results in DDX1-negative breast cancer patients treated with BCS and adjuvant RT.

FUNDING SOURCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conclusions
  7. FUNDING SOURCES
  8. REFERENCES

B.G.H. was supported by The Breast Cancer Research Foundation.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conclusions
  7. FUNDING SOURCES
  8. REFERENCES