Abstract
Background:
Leukemia and lymphoma are cancers affecting children, adolescents, and young adults and may affect reproductive outcomes and maternal metabolism. We evaluated for metabolic changes in newborns of mothers with a history of these cancers.
Methods:
A cross-sectional study was conducted on California births from 2007 to 2011 with linked maternal hospital discharge records, birth certificate, and newborn screening metabolites. History of leukemia or lymphoma was determined using ICD-9-CM codes from hospital discharge data and newborn metabolite data from the newborn screening program.
Results:
A total of 2,068,038 women without cancer history and 906 with history of leukemia or lymphoma were included. After adjusting for differences in maternal age, infant sex, age at metabolite collection, gestational age, and birthweight, among newborns born to women with history of leukemia/lymphoma, several acylcarnitines were significantly (p<0.001 – based on Bonferroni correction for multiple testing) higher compared to newborns of mothers without cancer history: C3-DC (mean difference (MD) = 0.006), C5-DC (MD = 0.009), C8:1 (MD = 0.008), C14 (MD = 0.010), and C16:1 (MD = 0.011), whereas citrulline levels were significantly lower (MD = −0.581) among newborns born to mothers with history of leukemia or lymphoma compared to newborns of mothers without a history of cancer.
Conclusion:
The varied metabolite levels suggest history of leukemia or lymphoma has metabolic impact on newborn offspring, which may have implications for future metabolic consequences such as necrotizing enterocolitis and urea cycle enzyme disorders in children born to mothers with a history of leukemia or lymphoma.
Keywords: leukemia, lymphoma, metabolite, newborn, amino acid, acylcarnitines
INTRODUCTION
Leukemia and lymphoma are two of the top cancers affecting children, adolescents and young adults. Those who have a history of childhood cancers, such as leukemia or lymphoma, have been shown to develop metabolic disorders in adulthood, including dyslipidemia, diabetes, and metabolic syndrome. Several studies have shown that a history of leukemia increases risk of dyslipidemia in patients in their adulthood, years after treatment cessation [1–6]. A study by Meacham et al., they found that long-term childhood cancer survivors were 1.6 times (95% CI 1.2–2.2) more likely to have diabetes compared to siblings without cancer, and additionally those exposed to abdominal irradiation or treated with an alkylating agent, corticosteroid, or anthracycline had higher prevalence of diabetes [7]. Metabolic syndrome is comprised of risk factors for cardiovascular diseases, which include hyperlipidemia, obesity, hypertension and insulin resistance. A meta-analysis has found that cancer survivors, including those with a history of leukemia and non-Hodgkin’s lymphoma, have an increased risk of metabolic syndrome [8]. Metabolic disorders are consistently associated with altered metabolite levels. For instance, studies have shown altered metabolite levels of amino acids among those with diabetes compared to those without diabetes [9,10].
Many of the metabolic disorders have underlying metabolic disturbances in fatty acid β-oxidation [7,9]. Fatty acid β-oxidation (FAO) is crucial in cancer cell function and survival, and mitochondrial fatty and amino acid β-oxidation metabolites are altered in patients with cancer [11]. Free fatty acids (FFAs) generate reactive oxygen species to induce oxidative stress and help FAO to produce energy for the survival of cancer cells [11,12]. Studies have shown that serum FFAs are altered in patients diagnosed with cancer compared to healthy individuals and that altered serum metabolites are present in patients with cancer prior to any treatments [13–20]. Specifically, studies assessing amino acid levels prior to any type of surgery or treatment in breast cancer, oral cancer, and lung cancer patients found that the levels differed from healthy patients with no cancer history [15–18,20]. Additionally, compared to healthy individuals, cancer patients who have undergone treatment have altered serum FFAs [21,22].
As metabolic disorders are seen in adulthood after having prior history of childhood cancers, the persistence of altered metabolites among pregnant women with a history of cancer may result in increased metabolic vulnerability in the newborn. Metabolic vulnerability can be as defined as altered metabolite levels that predict further health complications in newborns including necrotizing enterocolitis [23]. Previous research has demonstrated that a newborn’s metabolite profile is representative of their mother’s profile and newborn metabolic profiles of mitochondrial fatty and amino acid β-oxidation metabolites are markers of the metabolic vulnerability of the newborn [24]. Newborn metabolic screening (NBS) is a type of screening conducted within the first few hours to few days of life using tandem mass spectrometry. This screening captures a variety of biomarkers that represent mitochondrial FAO pathways. Altered metabolites, measured during newborn screening have been previously associated by our group and others with life-threatening serious newborn complications. These complications include necrotizing enterocolitis (NEC), respiratory distress syndrome, persistent pulmonary hypertension and patent ductus arteriosus [23,25,26]. Term infants with NEC have been shown to have increased levels of fatty acids in their NBS compared to infants without NEC [27]. Infants with persistent pulmonary hypertension were found to have high levels of phenylalanine and low levels of tyrosine levels, which are amino acids captured during NBS [25]. Taken together, these data suggest that the metabolic profiles measured as part of NBS can represent metabolic vulnerability of a newborn.
We hypothesized that altered FAO in the mother due to cancer and its treatment may increase metabolic vulnerability in the newborn, as represented by the metabolic profile. In this study, we evaluated whether a maternal diagnosis history of leukemia or lymphoma was associated with metabolic changes in newborns. Leukemias and lymphomas are among the most frequent cancers affecting children, adolescents and young adults. Given that these cancers are highly curable, with approximately 90% five-year survival rates for most types, there is much concern over reproductive outcomes and offspring well-being [28,29].
METHODS
Data Source and Linkages
We obtained a linked dataset maintained by the California Office of Statewide Health Planning and Development (OSHPD) that contains information on all births from 2007–2011. The dataset contains linked hospital discharge, birth certificate and death data from birth to one year of age and is further linked with newborn screening program data containing metabolite information. The hospital discharge data contains diagnoses and procedure codes that are based on the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM).
Study Population
There were 2,656,480 live born infants between 2007 and 2011. We included all mother-child pairs for live births between 20 and 44 weeks in which both mother and baby were linked to hospital discharge records and further linked to newborn screening data. Mothers who were <45 years of age at delivery and who were diagnosed with either leukemia or lymphoma or had no history of any type of cancer were included. Exclusion criteria were incomplete metabolite data, newborn screening collected at 48 hours or later and women with other types of cancer.
Study Variables
The primary exposure was history of leukemia or lymphoma ascertained from the hospital discharge record. The following commonly used ICD-9 codes were used to identify leukemia and lymphoma diagnoses: 201.x-202.x, 203.1x, 204.x-208.x, V10.6, and V10.7.
The primary outcomes were the metabolites (N=42) obtained from the newborn screening results. We included all metabolites in the California newborn screening panel. We assessed 12 amino acids (alanine, arginine, citrulline, isoleucine or leucine, glycine, 5-oxoproline, ornithine, proline, methionine, phenylalanine, tyrosine, valine), free carnitine, which is derived from the amino acid lysine and acts as a transporter of long-chain fatty acids into the mitochondria to be oxidized, 26 acylcarnitines, which are intermediates in fatty acid and amino acid breakdown, (C2 (acetylcarnitine), C3 (propionylcarnitine), C3-DC (malonylcarnitine), C4 (butyrylcarnitine+isobutyrylcarnitine), C5 (isovalerylcarnitine+methylbutyrylcarnitine), C5:1 (tiglylcarnitine), C5-DC (glutarylcarnitine), C5-OH (3-hydroxyisovalerylcarnitine), C6 (hexanoylcarnitine), C8 (octanoylcarnitine), C8:1 (octenoylcarnitine), C10 (decanoylcarnitine), C10:1 (decenoylcarnitine), C12 (dodecanoylcarnitine), C12:1 (dodecenoylcarnitine), C14 (tetradecanoylcarnitine), C14:1 (tetradecenoylcarnitine), C14-OH (3-hydroxytetradecanoylcarnitine), C16 (palmitoylcarnitine), C16:1 (palmitoleylcarnitine), C16-OH (3-hydroxypalmitoylcarnitine), C18 (stearoylcarnitine), C18:1 (oleoylcarnitine), C18-OH (3-hydroxystearoylcarnitine), C18:2 (linoleoylcarnitine), C18:1-OH (3-hydroxyoleoylcarnitine)), and three hormones/enzymes (thyroid stimulating hormone, 17-hydroxyprogesterone, Galactose-1-Phosphate Uridyl Transferase) [30–34].
The covariates assessed were obtained from birth certificates and newborn screening results. These covariates were selected based on their importance as seen in previous studies and the research question [23,24]. The covariates include maternal age, maternal race (non-Hispanic White, Asian, Black, Hispanic, or Other race), maternal education (<12 years, 12 years, and >12 years), smoking during pregnancy (yes/no), prior live births (0, 1, 2, and 3 or more), plurality (singleton and twins or more), sex of infant (male/female), age at newborn screening collection, gestational age, and birthweight. Maternal age, age at newborn screening collection, gestational age, and birthweight were all analyzed as continuous variables.
Statistical Analysis
For comparing descriptive characteristics, we used Chi-square tests and t-tests for categorical variables and continuous variables, respectively. The relationship of leukemia/lymphoma with each metabolite was evaluated using linear regression. We analyzed mean and standard deviation for each metabolite by the primary exposure. For unadjusted mean differences, we used an unadjusted regression model. A p-value <0.001 was used for assessing statistical significance after Bonferroni correction for multiple testing. We adjusted for potential confounders using multivariable linear regression models for the metabolites. All analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC).
The methods and protocols for the study were approved by the Committee for the Protection of Human Subjects within the Health and Human Services Agency of the State of California. De-identified data was provided to the researchers by the California Office of Statewide Health Planning and Development (Protocol # 12–09–0702) and determined not to qualify as human subjects research by the University of Iowa Institutional Review Board (IRB no.: 201602793).
RESULTS
A total of 2,068,944 women met study criteria. Among these women, 906 had a diagnosis code for leukemia or lymphoma on a hospital discharge abstract and 2,068,038 women had no indication of cancer. In Table 1, we see that women with a history of leukemia/lymphoma were on average one year older than those without cancer (29.2 years vs. 28.1 years, respectively), were more likely to have 12 or more years of education (60.4% vs 46.0%), and were more likely to be non-Hispanic White (44.0% vs 25.7%). Both the controls and those with a history of leukemia/lymphoma were mostly non-smokers during pregnancy (95.7%) and had newborns with a mean birthweight of approximately 3300 grams.
Table 1.
Descriptive characteristics of California women <45 years of age with leukemia/lymphoma who gave birth between 2007–2011
| Variable | Description | Total (n=2068944) | Leukemia/Lymphoma (n=906) | No Cancer (n=2068038) | p-value+ |
|---|---|---|---|---|---|
| Maternal age at delivery | <20 | 189369 (9.2%) | 59 (6.5%) | 189310 (9.2%) | <0.001 |
| 20–24 | 447248 (21.6%) | 158 (17.4%) | 447090 (21.6%) | ||
| 25–29 | 561403 (27.1%) | 233 (25.7%) | 561170 (27.1%) | ||
| 30–34 | 517942 (25.0%) | 266 (29.4%) | 517676 (25.0%) | ||
| 35–39 | 286363 (13.8%) | 154 (17.0%) | 286209 (13.8%) | ||
| 40–44 | 66619 (3.2%) | 36 (4.0%) | 66583 (3.2%) | ||
| Maternal age (continuous) | Mean and SD | 28.1 (6.2) | 29.2 (6.1) | 28.1 (6.2) | <0.001 |
| Median and IQR | 28.0 (23.0, 33.0) | 30.0 (25.0, 34.0) | 28.0 (23.0, 33.0) | ||
| Min and Max | (13.0, 44.0) | (13.0, 43.0) | (13.0, 44.0) | ||
| Maternal race | Asian | 261270 (12.6%) | 74 (8.2%) | 261196 (12.6%) | <0.001 |
| Black | 102671 (5.0%) | 49 (5.4%) | 102622 (5.0%) | ||
| Hispanic | 1028846 (49.7%) | 306 (33.8%) | 1028540 (49.7%) | ||
| Other race | 143308 (6.9%) | 78 (8.6%) | 143230 (6.9%) | ||
| Non-Hispanic White | 532849 (25.8%) | 399 (44.0%) | 532450 (25.7%) | ||
| Smoking history during pregnancy | No smoking | 1980960 (95.7%) | 851 (93.9%) | 1980109 (95.7%) | 0.007 |
| Smoked during pregnancy | 87984 (4.3%) | 55 (6.1%) | 87929 (4.3%) | ||
| Prior live births | 0 | 812865 (39.3%) | 406 (44.8%) | 812459 (39.3%) | 0.003 |
| 1 | 655903 (31.7%) | 278 (30.7%) | 655625 (31.7%) | ||
| 2 | 354813 (17.1%) | 138 (15.2%) | 354675 (17.2%) | ||
| 3 or more | 244113 (11.8%) | 84 (9.3%) | 244029 (11.8%) | ||
| Missing | 1250 (0.1%) | 1250 (0.1%) | |||
| Maternal education | <12 years | 516673 (25.0%) | 120 (13.2%) | 516553 (25.0%) | <0.001 |
| 12 years | 527814 (25.5%) | 199 (22.0%) | 527615 (25.5%) | ||
| >12 years | 952541 (46.0%) | 547 (60.4%) | 951994 (46.0%) | ||
| Missing | 71916 (3.5%) | 40 (4.4%) | 71876 (3.5%) | ||
| Prior preterm delivery | No previous preterm delivery | 1245039 (60.2%) | 494 (54.5%) | 1244545 (60.2%) | 0.206 |
| Had previous preterm delivery | 10948 (0.5%) | 7 (0.8%) | 10941 (0.5%) | ||
| Missing | 812957 (39.3%) | 405 (44.7%) | 812552 (39.3%) | ||
| Plurality | Singleton | 2063053 (99.7%) | 903 (99.7%) | 2062150 (99.7%) | 0.793 |
| Twins and more | 5891 (0.3%) | 3 (0.3%) | 5888 (0.3%) | ||
| Sex of Infant | Female | 1015309 (49.1%) | 443 (48.9%) | 1014866 (49.1%) | 0.915 |
| Male | 1053635 (50.9%) | 463 (51.1%) | 1053172 (50.9%) | ||
| Gestational Weeks | Mean and SD | 38.8 (1.5) | 38.7 (1.8) | 38.8 (1.5) | 0.002 |
| Median and IQR | 39.0 (38.0, 40.0) | 39.0 (38.0, 40.0) | 39.0 (38.0, 40.0) | ||
| Min and Max | (20.0, 44.0) | (27.0, 43.0) | (20.0, 44.0) | ||
| Age At Newborn Screening Collection (in hours) | Mean and SD | 27.5 (8.0) | 28.1 (8.3) | 27.5 (8.0) | 0.018 |
| Median and IQR | 26.0 (23.0, 33.0) | 26.5 (23.0, 33.0) | 26.0 (23.0, 33.0) | ||
| Min and Max | (12.0, 47.0) | (12.0, 47.0) | (12.0, 47.0) | ||
| Birthweight | Mean and SD | 3356.0 (490.3) | 3342.2 (562.9) | 3356.1 (490.3) | 0.395 |
| Median and IQR | 3355.0 (3062.0, 3660.0) | 3374.0 (3045.0, 3680.0) | 3355.0 (3062.0, 3660.0) | ||
| Min and Max | (198.0, 9999.0) | (1015.0, 5325.0) | (198.0, 9999.0) | ||
| Missing | 3 (0.0%) | 3 (0.0%) |
the missing values were not included in the p-value
Bold: p<0.05
The mean differences (MD) and standard deviations (SD) in metabolites between the newborns born to mothers with leukemia or lymphoma and those with no cancer are displayed in Table 2. Positive differences at the p<0.001 level were seen for the short-chain acylcarnitines C3-DC (MD (SD) = 0.007 (0.046)) and C5-DC (MD (SD) = 0.010 (0.072)), the medium-chain acylcarnitine,C8:1 (MD (SD) = 0.008 (0.059)), and two long-chain acylcarnitines, C14 (MD (SD) = 0.011 (0.088)), and C16:1 (MD (SD) = 0.011 (0.095)), among those newborns born to mothers with a history of leukemia/lymphoma compared to those newborns born to mothers without a cancer history. A negative mean difference was seen for the amino acid citrulline (MD (SD) = −0.633 (4.699)). These differences did not change appreciably after adjustment for maternal age, sex of the infant, age at collection of the newborn metabolite, gestational age, and birthweight (Table 2). Additionally, the models were also run adjusting for size for gestational age in place of birthweight and the results remained the same.
Table 2.
Mean and Standard Deviation of Newborn Metabolites by Leukemia/Lymphoma Taken From Newborn Screenings <48 hours After Birth of Newborns Born in California From 2007–2011
| Metabolite | Leukemia/Lymphoma (n=906) Mean | No Cancer (n=2068038) Mean | Unadjusted Mean Difference | Adjusted Mean Difference+ |
|---|---|---|---|---|
| Mean (SD) | Mean (SD) | Mean Difference (SD) | Mean Difference (SD) | |
| Free Carnitine | 34.955 (16.137) | 34.959 (15.924) | −0.004 (15.924) | 0.044 (15.813) |
| Acylcarnitines | ||||
| C2 | 25.589 (8.662) | 25.280 (8.063) | 0.309 (8.063) | 0.262 (7.978) |
| C3 | 1.916 (0.826) | 1.938 (0.807) | −0.022 (0.807) | −0.031 (0.790) |
| C3-DC | 0.112 (0.051) | 0.105 (0.046) | 0.007 (0.046) ** | 0.006 (0.046) ** |
| C4 | 0.287 (0.154) | 0.273 (0.140) | 0.014 (0.140)* | 0.013 (0.139)* |
| C5 | 0.132 (0.071) | 0.126 (0.068) | 0.006 (0.068)* | 0.005 (0.065)* |
| C5:1 | 0.029 (0.026) | 0.029 (0.027) | −0.000 (0.027) | −0.000 (0.027) |
| C5-DC | 0.158 (0.076) | 0.148 (0.072) | 0.010 (0.072) ** | 0.009 (0.072) ** |
| C5-OH | 0.195 (0.091) | 0.203 (0.095) | −0.008 (0.095)* | −0.007 (0.095)* |
| C6 | 0.072 (0.041) | 0.072 (0.045) | 0.000 (0.045) | −0.000 (0.045) |
| C8 | 0.078 (0.045) | 0.079 (0.101) | −0.001 (0.101) | −0.001 (0.101) |
| C8:1 | 0.127 (0.063) | 0.119 (0.059) | 0.008 (0.059) ** | 0.008 (0.059) ** |
| C10 | 0.098 (0.052) | 0.098 (0.050) | 0.000 (0.050) | 0.001 (0.050) |
| C10:1 | 0.060 (0.031) | 0.059 (0.032) | 0.001 (0.032) | 0.001 (0.032) |
| C12 | 0.180 (0.096) | 0.180 (0.097) | 0.000 (0.097) | 0.000 (0.097) |
| C12:1 | 0.107 (0.066) | 0.103 (0.062) | 0.004 (0.062)* | 0.004 (0.062)* |
| C14 | 0.251 (0.094) | 0.241 (0.088) | 0.011 (0.088) ** | 0.010 (0.087) ** |
| C14:1 | 0.154 (0.070) | 0.149 (0.071) | 0.005 (0.071)* | 0.004 (0.070) |
| C14-OH | 0.032 (0.025) | 0.031 (0.025) | 0.001 (0.025) | 0.001 (0.025) |
| C16 | 3.155 (1.051) | 3.068 (0.984) | 0.087 (0.985)* | 0.077 (0.967)* |
| C16:1 | 0.263 (0.105) | 0.251 (0.095) | 0.011 (0.095) ** | 0.011 (0.095) ** |
| C16-OH | 0.038 (0.025) | 0.038 (0.027) | 0.000 (0.027) | 0.000 (0.027) |
| C18 | 0.875 (0.300) | 0.855 (0.287) | 0.020 (0.287)* | 0.016 (0.285) |
| C18:1 | 1.212 (0.390) | 1.185 (0.371) | 0.027 (0.371)* | 0.021 (0.365) |
| C18-OH | 0.020 (0.018) | 0.020 (0.020) | −0.000 (0.020) | −0.000 (0.020) |
| C18:2 | 0.192 (0.096) | 0.194 (0.096) | −0.002 (0.096) | −0.003 (0.093) |
| C18:1-OH | 0.029 (0.020) | 0.028 (0.022) | 0.001 (0.022) | 0.001 (0.022) |
| Amino Acids | ||||
| 5-Oxoproline (OXP) | 36.152 (22.066) | 35.672 (19.540) | 0.481 (19.541) | 0.464 (19.531) |
| Alanine (ALA) | 268.275 (81.767) | 270.251 (85.488) | −1.976 (85.486) | −0.372 (84.229) |
| Arginine (ARG) | 9.961 (9.796) | 9.899 (7.649) | 0.062 (7.650) | −0.065 (7.573) |
| Citrulline (CIT) | 15.797 (4.554) | 16.430 (4.699) | −0.633 (4.699) ** | −0.581 (4.683) ** |
| Glycine (GLY) | 600.450 (168.207) | 594.172 (159.892) | 6.278 (159.896) | 6.478 (159.665) |
| Leucine or Isoleucine (XLE) | 103.857 (33.000) | 101.606 (28.923) | 2.251 (28.925)* | 1.593 (28.325) |
| Methionine (MET) | 27.364 (9.004) | 27.468 (8.650) | −0.104 (8.650) | −0.086 (8.443) |
| Ornithine (ORN) | 105.461 (38.399) | 105.365 (36.507) | 0.097 (36.507) | −0.339 (35.920) |
| Phenylalanine (PHE) | 63.770 (14.597) | 62.804 (14.237) | 0.966 (14.237)* | 1.090 (13.714)* |
| Proline (PRO) | 179.926 (58.062) | 179.707 (57.662) | 0.219 (57.662) | 0.069 (57.443) |
| Tyrosine (TYR) | 95.747 (39.395) | 98.625 (39.255) | −2.878 (39.255)* | −3.469 (38.345)* |
| Valine (VAL) | 106.985 (33.185) | 105.298 (32.198) | 1.687 (32.199) | 1.523 (32.137) |
| Hormones/Enzymes | ||||
| Thyroid Stimulating Hormone (TSH) | 5.607 (8.664) | 5.419 (5.331) | 0.188 (5.333) | 0.318 (5.139) |
| 17-Hydroxyprogesterone (OHP) | 24.655 (23.357) | 23.001 (17.886) | 1.654 (17.889)* | 1.338 (16.175)* |
| Galactose-1-Phosphate Uridyl Transferase (TRA) | 252.504 (48.947) | 254.991 (47.707) | −2.487 (47.707) | −2.282 (47.474) |
p<0.05
p<0.001
Adjusted for: maternal age, sex of the infant, age at newborn metabolite collection, gestational age, and birthweight
SD: Standard Deviation
DISCUSSION
To our knowledge, this study is the first to evaluate newborn screening data to assess if maternal cancer history of leukemia or lymphoma leads to metabolic vulnerability in newborns. This study is also the first to report statistically significant elevations in the acylcarnitines C3-DC, C5-DC, C8:1, C14, C16:1, and decrease in the amino acid citrulline in those newborns born to mothers with a history of leukemia or lymphoma and newborns born to mothers without a prior history of cancer. None of the other metabolites reached the p<0.001 threshold for a significant difference, correcting for multiple comparisons.
The increased levels of acyclcarntines among newborns born to mothers with a history of leukemia or lymphoma is consistent with other studies assessing acylcarnitine levels in cancer patients. Acylcarnitines are intermediates in the fatty acid and amino acid breakdown. As intermediates, they are transported into the mitochondria where the carnitines are removed from the acylcarnitines and regenerated to acyl-coenzyme A to be used for FAO [30,32]. A study by Qiu et al. found increased concentrations of acylcarnitines, which are FAO metabolites, in breast cancer patients compared to healthy individuals [35]. This study did not specify, however, if they were recent diagnoses or years after diagnosis and treatment. Additionally, in a study by Ke et al., the researchers collected plasma samples from patients in China with a history of epithelial ovarian cancer (EOC) who were not currently on any medications and used ultra-performance liquid chromatography mass spectrometry (UPLC/MS) to identify ovarian cancer-related metabolic signatures. The researchers found that compared to benign ovarian tumor and uterine fibroid patients, EOC patients had significantly increased plasma concentrations of acylcarnitines [36]. These increased levels of acylcarnitines indicate higher FAO in cancer patients [35,36].
Additionally, we found a reduced mean value of citrulline among newborns born to mothers with a history of leukemia or lymphoma compared to newborns born to mothers without a history of cancer. Although in our study, we did not have cancer treatment data such as type of treatment and timing of the treatment, possible explanations for the reduced citrulline levels include effects of cancer treatments or effects of pregnancy itself. Citrulline is an amino acid produced from arginine during nitric oxide (NO) production. The depletion of arginine is an important therapeutic strategy that is used in cancer treatment, including the treatments for leukemia and lymphoma [37–41]. NO-mediated angiogenesis promotes tumor growth and survival, thus depleting arginine and its metabolites causes an attenuation of tumor growth [37–42]. This leads to low levels of essential amino acids and nutrients in the body (i.e., a low resource setting) [43]. Those treated with chemotherapy such as leukemia and lymphoma patients are prone to gut dysbiosis, which is a microbial imbalance in the gastrointestinal tract [44]. This gut dysbiosis also causes a low-resource setting [43,44]. Cancer history may contribute to the low-resource setting associated with pregnancy, with minimal amino acids and other metabolites [43]. With the majority of citrulline being used by the mother during treatment and having reduced levels of essential amino acids, it is possible that this leads to the low levels in their newborns.
Research has already shown that a newborn’s metabolite profile is representative of their mother’s profile [24]. Studies have shown that there is an exchange of amino acids between the human fetus and placenta [45,46]. Therefore, it is possible that arginine metabolism is intrinsically modulated by the adaptive immune mechanisms in leukemia or lymphoma patients and the exchange of amino acids between the mother and fetus results in lower levels of arginine metabolites, including citrulline, in the newborns born to mothers with leukemia/lymphoma compared to newborns born to mothers without a history of cancer.
Our study had a few limitations. We used ICD-9-CM codes including V10 history codes to determine our comparison groups of leukemia or lymphoma and no cancer. Under-identification of cancer history could have resulted in misclassification. However, because the exposure of leukemia and lymphoma is rare, the impact of low sensitivity on the results would not be great. Additionally, we were unable to distinguish the time between leukemia/lymphoma diagnosis and the birth, so we are unable to determine whether our findings reflect a sustained metabolic effect of leukemia or lymphoma or an effect of recent cancer diagnosis and treatment. However, 56.7% of the women with leukemia/lymphoma were ascertained via a V10 history code, suggesting that these women were primarily cancer survivors and not in active treatment. Further, from a previous study, we examined the statewide Iowa Surveillance, Epidemiology, and End Results (SEER) registry with linked birth certificates and found that the time from diagnosis to birth ranged from less than 1 year to 37 years with only 20% of diagnoses occurring within 3 years of delivery (data not shown). We also were not able to distinguish between the different types of leukemias and lymphomas such as Hodgkin’s lymphoma vs. Non-Hodgkin’s lymphoma and acute lymphoblastic leukemia vs. acute myelocytic leukemia. Additionally, we did not have information on the mother’s treatment of cancer to evaluate metabolite levels in the newborn based on the different treatments that the mother received for cancer. From the aforementioned Iowa SEER data, 49% of mothers with a history of leukemia or lymphoma were treated with only chemotherapy and 34.2% were treated with both chemotherapy and radiation.
The study also had several strengths. The sample size was large and the study population was racially and ethnically diverse, providing strong generalizability. We also used newborn screening data to assess if maternal cancer history of leukemia or lymphoma leads to metabolic vulnerability in newborns, which, to our knowledge, has not previously been studied. Furthermore, we were able to examine a wide range of metabolites including amino acids, free carnitine, and acylcarnitines.
In conclusion, we found increased mean levels of specific acylcarnitines and decreased level of citrulline in the newborn screening results of newborns born to mothers with a history of leukemia or lymphoma compared to the newborns of mothers without a history of cancer. While the risk differences for many of these metabolites were small, similar differences have been associated with neonatal complications including necrotizing enterocolitis (NEC), respiratory distress syndrome, persistent pulmonary hypertension and patent ductus arteriosus [23,25,26]. For example, newborns with low citrulline levels have been shown to have respiratory distress syndrome and other urea cycle enzyme disorders [47,48]. On the other hand, elevations in acylcarnitines increase the risk for necrotizing enterocolitis and fatty acid oxidation disorders [23,49,50]. The metabolic differences identified would likely not affect clinical decision-making but may indicate biologically relevant targets for future studies. Further studies are warranted to assess the metabolite levels of newborns born to those with a cancer history and then follow these newborns with altered metabolite levels for several years to assess long-term health complications and any further changes in their metabolites. Additional studies assessing persistence of metabolite abnormalities among cancer survivors and metabolite studies accounting for time since cancer diagnosis would also make valuable contributions to this field. Our study demonstrates the potential need for monitoring of newborn metabolites involved in FAO, such as carnitines and amino acids, prior to discharge of the newborn. This would enable clinicians to use this information to initiate mechanisms for early diagnosis, treatment, and nutritional supplementation options tailored to an individual neonate’s metabolic profile. This will also help to reduce morbidity and mortality caused by health complications from altered metabolites in this population.
Acknowledgments
Declaration of Interest
This research is supported by the National Cancer Institute (P30 CA086862-18S6). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health. Dr. Kelli Ryckman and Dr. Sonia Anand received the grant during the conduct of the study. Additionally, Dr. Ryckman has a patent for Serum Screening and Lipid Markers Predicting Preterm Birth pending. All other authors report no conflict of interest. Declaration of Interest
Reference:
- 1.Barbosa-Cortés L, López-Alarcón M, Mejía-Aranguré JM, et al. Adipokines, insulin resistance, and adiposity as a predictors of metabolic syndrome in child survivors of lymphoma and acute lymphoblastic leukemia of a developing country. BMC cancer. 2017. 2017/02/13;17(1):125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Barnea D, Raghunathan N, Friedman DN, et al. Obesity and Metabolic Disease After Childhood Cancer. Oncology (Williston Park, NY). 2015;29(11):849–855. [PMC free article] [PubMed] [Google Scholar]
- 3.Fournier M, Bonneil E, Garofalo C, et al. Altered proteome of high-density lipoproteins from paediatric acute lymphoblastic leukemia survivors. Scientific reports. 2019. 2019/03/12;9(1):4268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Karakaya P, Yılmaz S, Tüfekçi O, et al. Endocrinological and cardiological late effects among survivors of childhood acute lymphoblastic leukemia. Turkish journal of haematology : official journal of Turkish Society of Haematology. 2013;30(3):290–299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Malhotra J, Tonorezos ES, Rozenberg M, et al. Atherogenic low density lipoprotein phenotype in long-term survivors of childhood acute lymphoblastic leukemia. J Lipid Res. 2012. Dec;53(12):2747–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Morel S, Leahy J, Fournier M, et al. Lipid and lipoprotein abnormalities in acute lymphoblastic leukemia survivors. J Lipid Res. 2017;58(5):982–993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Meacham LR, Sklar CA, Li S, et al. Diabetes mellitus in long-term survivors of childhood cancer. Increased risk associated with radiation therapy: a report for the childhood cancer survivor study. Archives of internal medicine. 2009. Aug 10;169(15):1381–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jung H-S, Myung S-K, Kim B-S, et al. Metabolic syndrome in adult cancer survivors: A meta-analysis. Diabetes Research and Clinical Practice. 2012. 2012/02/01/;95(2):275–282. [DOI] [PubMed] [Google Scholar]
- 9.Yang SJ, Kwak S-Y, Jo G, et al. Serum metabolite profile associated with incident type 2 diabetes in Koreans: findings from the Korean Genome and Epidemiology Study. Scientific reports. 2018. 2018/05/29;8(1):8207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Yun JH, Lee H-S, Yu H-Y, et al. Metabolomics profiles associated with HbA1c levels in patients with type 2 diabetes. PloS one. 2019;14(11):e0224274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Carracedo A, Cantley LC, Pandolfi PP. Cancer metabolism: fatty acid oxidation in the limelight. Nature reviews Cancer. 2013. Apr;13(4):227–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Xiong J. Fatty Acid Oxidation in Cell Fate Determination. Trends in biochemical sciences. 2018. May 4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Heber D, Byerly LO, Chlebowski RT. Metabolic abnormalities in the cancer patient. Cancer. 1985. Jan 1;55(1 Suppl):225–9. [DOI] [PubMed] [Google Scholar]
- 14.Lai HS, Lee JC, Lee PH, et al. Plasma free amino acid profile in cancer patients. Semin Cancer Biol. 2005. Aug;15(4):267–76. [DOI] [PubMed] [Google Scholar]
- 15.Miyagi Y, Higashiyama M, Gochi A, et al. Plasma free amino acid profiling of five types of cancer patients and its application for early detection. PloS one. 2011;6(9):e24143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Proenza AM, Oliver J, Palou A, et al. Breast and lung cancer are associated with a decrease in blood cell amino acid content. The Journal of nutritional biochemistry. 2003. Mar;14(3):133–8. [DOI] [PubMed] [Google Scholar]
- 17.Tiziani S, Lopes V, Gunther UL. Early stage diagnosis of oral cancer using 1H NMR-based metabolomics. Neoplasia. 2009. Mar;11(3):269–76, 4p following 269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cui M, Wang Q, Chen G. Serum metabolomics analysis reveals changes in signaling lipids in breast cancer patients. Biomedical chromatography : BMC. 2016. Jan;30(1):42–7. [DOI] [PubMed] [Google Scholar]
- 19.Lv W, Yang T. Identification of possible biomarkers for breast cancer from free fatty acid profiles determined by GC-MS and multivariate statistical analysis. Clin Biochem. 2012. Jan;45(1–2):127–33. [DOI] [PubMed] [Google Scholar]
- 20.Mazzone PJ, Wang XF, Beukemann M, et al. Metabolite Profiles of the Serum of Patients with Non-Small Cell Carcinoma. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2016. Jan;11(1):72–8. [DOI] [PubMed] [Google Scholar]
- 21.Zhang Y, Song L, Liu N, et al. Decreased serum levels of free fatty acids are associated with breast cancer. Clinica chimica acta; international journal of clinical chemistry. 2014. Nov 1;437:31–7. [DOI] [PubMed] [Google Scholar]
- 22.Liu J, Mazzone PJ, Cata JP, et al. Serum free fatty acid biomarkers of lung cancer. Chest. 2014. Sep;146(3):670–679. [DOI] [PubMed] [Google Scholar]
- 23.Sylvester KG, Kastenberg ZJ, Moss RL, et al. Acylcarnitine Profiles Reflect Metabolic Vulnerability for Necrotizing Enterocolitis in Newborns Born Premature. The Journal of pediatrics. 2017. Feb;181:80–85.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ryckman KK, Shchelochkov OA, Cook DE, et al. The influence of maternal disease on metabolites measured as part of newborn screening. The journal of maternal-fetal & neonatal medicine : the official journal of the European Association of Perinatal Medicine, the Federation of Asia and Oceania Perinatal Societies, the International Society of Perinatal Obstet. 2013. Sep;26(14):1380–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kaluarachchi DC, Smith CJ, Klein JM, et al. Polymorphisms in urea cycle enzyme genes are associated with persistent pulmonary hypertension of the newborn. Pediatric research. 2018. Jan;83(1–1):142–147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sood MM, Murphy MSQ, Hawken S, et al. Association Between Newborn Metabolic Profiles and Pediatric Kidney Disease. Kidney international reports. 2018. May;3(3):691–700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lin J. Too much short chain fatty acids cause neonatal necrotizing enterocolitis. Medical hypotheses. 2004. 2004/02/01/;62(2):291–293. [DOI] [PubMed] [Google Scholar]
- 28.SEER NCI. Fast Stats 2018. [cited 2018 05-01-2018]. Available from: https://seer.cancer.gov/statfacts/
- 29.Hudson MM. Reproductive outcomes for survivors of childhood cancer. Obstetrics and gynecology. 2010. Nov;116(5):1171–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Jakobs BS, Wanders RJ. Fatty acid beta-oxidation in peroxisomes and mitochondria: the first, unequivocal evidence for the involvement of carnitine in shuttling propionyl-CoA from peroxisomes to mitochondria. Biochemical and biophysical research communications. 1995. Aug 24;213(3):1035–41. [DOI] [PubMed] [Google Scholar]
- 31.Peluso G, Nicolai R, Reda E, et al. Cancer and anticancer therapy-induced modifications on metabolism mediated by carnitine system. Journal of cellular physiology. 2000. Mar;182(3):339–50. [DOI] [PubMed] [Google Scholar]
- 32.Ramsay RR, Gandour RD, van der Leij FR. Molecular enzymology of carnitine transfer and transport. Biochimica et biophysica acta. 2001. Mar 9;1546(1):21–43. [DOI] [PubMed] [Google Scholar]
- 33.Rogalidou M, Evangeliou A, Stiakaki E, et al. Serum Carnitine Levels in Childhood Leukemia. Journal of pediatric hematology/oncology. 2010;32(2). [DOI] [PubMed] [Google Scholar]
- 34.Yaris N, Akyuz C, Coskun T, et al. Serum carnitine levels of pediatric cancer patients. Pediatric hematology and oncology. 2002. Jan-Feb;19(1):1–8. [DOI] [PubMed] [Google Scholar]
- 35.Qiu Y, Zhou B, Su M, et al. Mass spectrometry-based quantitative metabolomics revealed a distinct lipid profile in breast cancer patients. International journal of molecular sciences. 2013. Apr 12;14(4):8047–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ke C, Hou Y, Zhang H, et al. Large-scale profiling of metabolic dysregulation in ovarian cancer. International journal of cancer. 2015. Feb 1;136(3):516–26. [DOI] [PubMed] [Google Scholar]
- 37.Barzal JA, Szczylik C, Rzepecki P, et al. Plasma citrulline level as a biomarker for cancer therapy-induced small bowel mucosal damage. Acta Biochim Pol. 2014;61(4):615–31. [PubMed] [Google Scholar]
- 38.Feun L, You M, Wu CJ, et al. Arginine deprivation as a targeted therapy for cancer. Current pharmaceutical design. 2008;14(11):1049–1057. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Kim S-H, Roszik J, Grimm EA, et al. Impact of l-Arginine Metabolism on Immune Response and Anticancer Immunotherapy. Frontiers in oncology. 2018;8:67–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Rashkovan M, Ferrando A. Metabolic dependencies and vulnerabilities in leukemia. Genes & development. 2019;33(21–22):1460–1474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Zou S, Wang X, Liu P, et al. Arginine metabolism and deprivation in cancer therapy. Biomedicine & Pharmacotherapy. 2019. 2019/10/01/;118:109210. [DOI] [PubMed] [Google Scholar]
- 42.Patil MD, Bhaumik J, Babykutty S, et al. Arginine dependence of tumor cells: targeting a chink in cancer’s armor. Oncogene. 2016;35(38):4957–4972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Weckman AM, McDonald CR, Baxter JB, et al. Perspective: L-arginine and L-citrulline Supplementation in Pregnancy: A Potential Strategy to Improve Birth Outcomes in Low-Resource Settings. Adv Nutr. 2019. Sep 1;10(5):765–777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Rashidi A, Kaiser T, Graiziger C, et al. Gut dysbiosis during antileukemia chemotherapy versus allogeneic hematopoietic cell transplantation. Cancer. 2020. Apr 1;126(7):1434–1447. [DOI] [PubMed] [Google Scholar]
- 45.Holm MB, Bastani NE, Holme AM, et al. Uptake and release of amino acids in the fetal-placental unit in human pregnancies. PloS one. 2017;12(10):e0185760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Jansson T. Amino acid transporters in the human placenta. Pediatric research. 2001. Feb;49(2):141–7. [DOI] [PubMed] [Google Scholar]
- 47.Endo F, Matsuura T, Yanagita K, et al. Clinical Manifestations of Inborn Errors of the Urea Cycle and Related Metabolic Disorders during Childhood. The Journal of nutrition. 2004;134(6):1605S–1609S. [DOI] [PubMed] [Google Scholar]
- 48.Ware LB, Magarik JA, Wickersham N, et al. Low plasma citrulline levels are associated with acute respiratory distress syndrome in patients with severe sepsis. Critical care (London, England). 2013;17(1):R10–R10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Rinaldo P, Cowan TM, Matern D. Acylcarnitine profile analysis. Genetics in Medicine. 2008. 2008/02/01;10(2):151–156. [DOI] [PubMed] [Google Scholar]
- 50.Yoon H-R. Screening newborns for metabolic disorders based on targeted metabolomics using tandem mass spectrometry. Ann Pediatr Endocrinol Metab. 2015;20(3):119–124. [DOI] [PMC free article] [PubMed] [Google Scholar]