Non-Hodgkin lymphomas (NHL) are blood cancers that originate from host immune cells. Established risk factors include human immunodeficiency virus (HIV) infection, organ transplantation, inherited immunodeficiency syndromes or autoimmune conditions. Infections that affect immune response, either by transforming immune cells or causing chronic immune stimulation are also established NHL risk factors.
Pregnancy-related factors are also known to affect immune response. It is thought that the immune tolerance that enables a woman to establish and maintain a successful pregnancy could potentially affect subsequent NHL risk. It is posited that increasing parity could similarly result in immune modulation that contributes to subsequent NHL. Progesterone expressed during pregnancy is also thought to be anti-inflammatory and could potentially reduce NHL risk.
The roles that pregnancy-related factors play in NHL development remain unclear. Here, we examine the relationship between pregnancy-related factors and risk of B-cell NHL among women in the Los Angeles County NHL Case-Control Study. Pregnancy-related factors evaluated included pregnancy history [ever (full-term) pregnant, number of full-term pregnancies], breast-feeding (ever breast-fed, duration, number of children breast-fed), nausea during pregnancy that required treatment or hospitalization, diethylstilbestrol (DES) use during pregnancy and lactation suppressant use after full-term pregnancy. In addition to overall B-cell NHL, major subtypes were further evaluated for risk, including diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), chronic lymphocytic leukaemia/small lymphocytic lymphoma (CLL/SLL) and marginal zone lymphoma (MZL).
The LA County NHL Case-Control Study has been described previously (Wang et al, 2017). Briefly, eligible case patients, identified by the LA County Cancer Surveillance Program, were diagnosed with primary B-cell NHL between the ages of 20 and 79 years from 1 May 2004 to 31 March 2008, resided in LA County at diagnosis, and had no history of any haematological malignancy or HIV before their diagnoses. The World Health Organization classification system was applied to classify NHL subtypes (Jaffe, 2009). Women who were residents of LA County with no history of haematological malignancy or HIV were individually matched to NHL patients on age, race, and socioeconomic status. A total of 997 matched pairs were recruited. Information regarding pregnancy-related factors and covariates was ascertained up to the case patient’s date of diagnosis, for each matched pair.
For all B-cell NHL, conditional logistic regression models were fit to estimate multivariable-adjusted odds ratios (OR) and 95% confidence intervals (CI) associated with pregnancy-related factors. For NHL subtypes, unconditional logistic regression models were fit to maximize study power. Covariates included in each model are provided in footnotes to each table, and their distribution shown in Table SI.
Compared to women who had never been pregnant, those who had ever been pregnant had a 27% lower risk of B-cell NHL (OR = 0·73, 95% CI = 0·56–0·96, Table I), and those reporting one or more full-term pregnancies had a 29% lower risk of B-cell NHL (OR = 0·71, 95% CI = 0·54–0·93). Increasing number of full-term pregnancies was also inversely associated with risk for B-cell NHL overall, DLBCL and FL (all P-trends ≤0·003). We note that significant associations for DLBCL and FL may be attributable to their larger sample sizes compared to other NHL subtypes. Nevertheless, our study results for number of full-term pregnancies are consistent with results from a Connecticut case-control study (Zhang et al, 2004) and the California Teachers Study (CTS) cohort (Prescott et al, 2009). Data from the recent international pooled analysis of case-control studies also supports decreased FL risk with increasing number of pregnancies (Kane et al, 2012).
Table I.
Odds ratios (OR) and 95% confidence intervals (CI) for associations between reproductive factors and B-cell NHL and NHL subtypes among women aged 20–79 years, living in Los Angeles County.
| Conditional analysis* | Unconditional analysis† | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Reproductive factors |
Cases (n = 997) |
Controls (n = 997) |
OR (95% CI)* | DLBCL (n = 278) |
OR (95% CI)† | FL (n = 241) |
OR (95% CI)† | CLL/SCLL (n = 214) |
OR (95% CI)† | MZL (n = 128) |
OR (95% CI)† |
| Ever pregnant | |||||||||||
| No§ | 165 | 142 | 1 | 56 | 1 | 43 | 1 | 31 | 1 | 15 | 1 |
| Yes | 832 | 855 | 0·73 (0·56–0·96) | 222 | 0·61 (0·42–0·88) | 198 | 0·71 (0·48–1·06) | 183 | 0·72 (0·46–1·13) | 113 | 0·97 (0·53–1·78) |
| Had full term (>26 week) pregnancy | |||||||||||
| No | 165 | 142 | 1 | 56 | 1 | 43 | 1 | 31 | 1 | 15 | 1 |
| Only short-term pregnancy | 78 | 72 | 0·91 (0·60–1·38) | 23 | 0·88 (0·49–1·57) | 18 | 0·79 (0·42–1·50) | 16 | 0·98 (0·49–1·96) | 9 | 1·15 (0·47–2·81) |
| Yes | 754 | 783 | 0·71 (0·54–0·93) | 199 | 0·57 (0·39–0·84) | 180 | 0·70 (0·47–1·05) | 167 | 0·69 (0·43–1·09) | 104 | 0·95 (0·52–1·75) |
| Number of full-term pregnancies | |||||||||||
| Never pregnant | 165 | 142 | 1 | 56 | 1 | 43 | 1 | 31 | 1 | 15 | 1 |
| 1 | 149 | 117 | 0·97 (0·69–1·37) | 46 | 0·94 (0·58–1·52) | 34 | 0·92 (0·54–1·56) | 28 | 0·90 (0·50–1·62) | 16 | 1·15 (0·53–2·49) |
| 2 | 261 | 290 | 0·67 (0·49–0·90) | 57 | 0·45 (0·29–0·71) | 70 | 0·71 (0·46–1·12) | 55 | 0·64 (0·39–1·07) | 43 | 1·14 (0·60–2·19) |
| 3 | 187 | 206 | 0·65 (0·46–0·90) | 52 | 0·55 (0·34–0·88) | 54 | 0·74 (0·46–1·20) | 36 | 0·56 (0·32–0·99) | 20 | 0·68 (0·32–1·42) |
| 4+ | 157 | 170 | 0·56 (0·39–0·80) | 44 | 0·48 (0·28–0·81) | 22 | 0·34 (0·19–0·63) | 48 | 0·76 (0·43–1·34) | 25 | 0·72 (0·33–1·55) |
| P-trend | 0·0003 | 0·0003 | 0·003 | 0·12 | 0·34 | ||||||
| Age at first full term pregnancy (years) | |||||||||||
| Never pregnant | 165 | 142 | 1 | 56 | 1 | 43 | 1 | 31 | 1 | 15 | 1 |
| <25 | 451 | 463 | 0·65 (0·48–0·87) | 106 | 0·42 (0·28–0·65) | 111 | 0·70 (0·45–1·08) | 114 | 0·74 (0·45–1·20) | 66 | 0·89 (0·46–1·71) |
| ≥25 | 302 | 320 | 0·76 (0·57–1·02) | 93 | 0·74 (0·49–1·10) | 69 | 0·70 (0·45–1·09) | 53 | 0·63 (0·38–1·05) | 38 | 1·02 (0·53–1 ·97) |
| Ever breastfed | |||||||||||
| No | 262 | 232 | 1 | 64 | 1 | 65 | 1 | 65 | 1 | 35 | 1 |
| Yes | 492 | 551 | 0·86 (0·69–1·09) | 135 | 0·95 (0·66–1·35) | 115 | 0·78 (0·55–1·12) | 102 | 0·81 (0·57–1·17) | 69 | 0·97 (0·61–1·53) |
| Breast feeding duration (weeks) | |||||||||||
| Never breastfed | 262 | 232 | 1 | 64 | 1 | 65 | 1 | 65 | 1 | 35 | 1 |
| <20 | 184 | 191 | 0·91 (0·68–1·21) | 48 | 0·94 (0·61–1·45) | 37 | 0·72 (0·45–1·14) | 45 | 0·98 (0·63–1·52) | 22 | 0·84 (0·47–1·51) |
| 20–59 | 159 | 170 | 0·92 (0·68–1·24) | 42 | 1·02 (0·64–1 ·61) | 42 | 0·94 (0·59–1·47) | 30 | 0·78 (0·48–1·28) | 23 | 1·18 (0·65–2·13) |
| ≥60 | 147 | 188 | 0·75 (0·55–1·02) | 44 | 0·88 (0·56–1·39) | 36 | 0·73 (0·45–1·16) | 26 | 0·64 (0·38–1·07) | 24 | 0·96 (0·54–1·72) |
| P-trend‡ | 0·08 | 0·72 | 0·23 | 0·03 | 0·88 | ||||||
| Number of children breast-fed (limited to parous women)‡ | |||||||||||
| 0 | 262 | 232 | 64 | 65 | 65 | 35 | |||||
| 1 | 193 | 196 | 0·98 (0·73–1·30) | 54 | 1 ·11 (0·72–1 ·71) | 46 | 0·91 (0·59–1·42) | 43 | 0·95 (0·61–1·49) | 18 | 0·74 (0·40–1·39) |
| 2 | 174 | 191 | 0·93 (0·69–1·24) | 40 | 0·84 (0·53–1·33) | 41 | 0·79 (0·50–1·25) | 36 | 0·87 (0·54–1·39) | 33 | 1·47 (0·85–2·54) |
| 3+ | 125 | 164 | 0·68 (0·49–0·93) | 41 | 0·88 (0·55–1·39) | 28 | 0·63 (0·38–1·04) | 23 | 0·60 (0·35–1·02) | 18 | 0·75 (0·40–1·40) |
| P-trend‡ | 0·03 | 0·41 | 0·06 | 0·03 | 0·95 | ||||||
NHL, non-Hodgkin lymphoma; DLBCL, Diffuse large B-cell Lymphoma; FL, Follicular Lymphoma; CLL/SLL, Chronic Lymphocytic Leukaemia/Small Lymphocytic Lymphoma; MZL, Marginal Zone Lymphoma.
Adjusted for education (less than high school, high school graduate, some college/technical school, college graduate); alcohol consumption status (never, former, current); smoking status (never, former, current); first degree family history of haematological malignancy (no, yes, unknown).
Adjusted for factors listed above; additionally adjusted for race (Non-Hispanic white, Black, Hispanic, Asian/others) and age (<50, 50–64, 65+ years).
Unconditional analysis.
Four women (1 case, 3 controls) were pregnant on their reference dates and are included in never pregnant category.
Decreased risk associations were also observed in our study for exposures related to increasing number of full-term pregnancies. Women who had their first full-term pregnancy before age 25 years had reduced B-cell NHL risk (OR = 0·65, 95% CI = 0·48–0·87), as did parous women who breast-fed at least three children (versus those who never breast-fed) (OR = 0·68, 95% CI = 0·49–0·93). These results are consistent with some previously published articles (Olsson et al, 1990; Cerhan et al, 2002) but not all (Frisch et al, 2006).
Among pregnancy-related treatments (Table II), our data suggest that treatment for severe nausea during pregnancy was not associated with risk of NHL in our study, which is consistent with results from an earlier study (Nelson et al, 2001). However, women who used DES during a pregnancy had two-fold greater risk of B-cell NHL overall than ever pregnant women who never used DES (OR = 2·37, 95% CI = 1·03–5·44) (Table II). Notably, use of DES during pregnancy was associated with five-fold greater risk of MZL (OR = 5·54, 95% CI = 1·76–17·45). This association is of potential interest as DES is known to alter immune function and is considered an oestrogenic endocrine disruptor. However, lactation suppressant use after full-term pregnancy was associated with decreased FL risk (OR = 0·61, 95% CI = 0·41–0·93), which concurs with Nelson et al (2001).
Table II.
Odds ratios (OR) and 95% confidence intervals (CI) for associations between pregnancy-related treatments and B-cell NHL and NHL subtypes among women aged 20–79 years, living in Los Angeles County.a
| Conditional analysis* | Unconditional analysis† | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Treatment information§ | Cases | Controls | OR (95% CI) | DLBCL | OR (95% CI) | FL | OR (95% CI) | CLL/SLL | OR (95% CI) | MZL | OR (95% CI) |
| Drug treatment or hospitalization for nausea during pregnancy | |||||||||||
| No | 709 | 753 | 1 | 195 | 1 | 175 | 1 | 151 | 1 | 92 | 1 |
| Yes | 106 | 90 | 1·19 (0·89–1·61) | 26 | 1·18 (0·73–1·90) | 21 | 1·00 (0·60–1·68) | 25 | 1·27 (0·78–2·08) | 19 | 1·65 (0·94–2·89) |
| Took DES during pregnancy | |||||||||||
| No | 778 | 813 | 1 | 210 | 1 | 189 | 1 | 170 | 1 | 102 | 1 |
| Yes | 18 | 10 | 2·37 (1·03–5·44)¶ | 4 | 2·55 (0·76–8·51) | 2 | 1·04 (0·22–4·92) | 5 | 2·09 (0·69–6·34) | 5 | 5·54 (1·76–17·45)†† |
| Used lactation suppressants after full team pregnancy (limited to parous women)‡ | |||||||||||
| No | 490 | 497 | 1 | 134 | 1 | 127 | 95 | 66 | |||
| Yes | 216 | 222 | 0·94 (0·74–1·20) | 53 | 0·86 (0·59–1·25) | 38 | 0·61 (0·41–0·93)** | 59 | 1·06 (0·72–1·55) | 35 | 1·06 (0·67–1·68) |
DLBCL, Diffuse large B-cell Lymphoma; FL, Follicular Lymphoma; CLL/SLL, Chronic Lymphocytic Leukaemia/Small Lymphocytic Lymphoma; MZL, Marginal Zone Lymphoma; DES, Diethylstilbestrol.
Adjusted for education (less than high school, high school graduate, some college/technical school, college/graduate); alcohol status 1 year before reference age (never, former, current); smoking status 1 year before reference age (never, former, current); 1st degree family history of haematological malignancy (no, yes, unknown).
Additionally adjusted for race (White, Black, Hispanic, Asian and other) and age (<50, 50–64, 65+ years).
Unconditional analysis.
Table excludes 165 cases (DLBCL 56, FL 43, CLL/SLL 31, and MZL 15) and 142 controls who were never pregnant. These women were included in the analyses but are not shown in this table (see Table I for risk estimates related to this group). Unknown group are not shown in this table.
P = 0.035.
P = 0.01.
P = 0.017.
We acknowledge that lymphomas occurring during pregnancy have been well-documented. It has been posited that the preponderance of lymphomas at the ovary and uterus could be due to increased blood flow to these organs or hormone receptors expressed during pregnancy (Evens et al, 2013). Although the reduced NHL risk associated with pregnancy in our study may appear to contradict those data, we note that the timing of exposures in these two scenarios is vastly different. The focus of our present study is on reproductive factors long before lymphoma develops.
In conclusion, our results provide evidence that increased parity, early age at first full-term pregnancy, and breast-feeding are associated with decreased risks of B-cell NHL. These associations are also observed with breast cancer, and warrant further investigation to uncover whether similar anti-carcinogenic mechanisms or specific hormone-related immune alterations are responsible. Our report is the first suggesting the association of DES use with increased B-cell NHL risk, particularly MZL, and warrants further investigation in collaborative efforts. Although no DES longer in use, should the association be replicated, it could provide important clues towards pinpointing biological mechanisms important for lymphomagenesis.
Supplementary Material
Table SI. Distribution of B-cell NHL in Los Angeles County NHL case and control study (2004–2008)1
Acknowledgement
LB and WC designed the research study; JL, YL, SSW, LB, JSH analysed the data and JL, YL, SSW, LB, JSH, WC wrote the paper.
Funding
This research was made possible through funding by the National Cancer Institute grants R01 CA166219, P01 CA017054, R01CA033572, and P30 CA033572.
Footnotes
Supporting Information
Additional supporting information may be found online in the Supporting Information section at the end of the article.
References
- Cerhan JR, Habermann TM, Vachon CM, Putnam SD, Zheng W, Potter JD & Folsom AR (2002) Menstrual and reproductive factors and risk of non-Hodgkin lymphoma: the Iowa women’s health study (United States). Cancer Causes & Control, 13, 131–136. [DOI] [PubMed] [Google Scholar]
- Evens AM, Advanci R, Press OW, Lossos IS, Vose JM, Hernandez-Ilizaliturri FJ, Robinson BK, Otis S, Nadav Dagan L, Abdallah R, Kroll-Desrosiers A, Yarber JL, Sandoval J, Foyil K, Parker LM, Gordon LI, Blum KA, Flowers CR, Leonard JP, Habermann TM & Bartlett NL (2013) Lymphoma occuring during pregnancy: antenatal therapy, complications, and maternal survival in a multicenter analysis. Journal of Clinical Oncology, 31, 4132–4139. [DOI] [PubMed] [Google Scholar]
- Frisch M, Pedersen BV, Wohlfahrt J, Hjalgrim H & Biggar RJ (2006) Reproductive patterns and non-Hodgkin lymphoma risk in Danish women and men. European Journal of Epidemiology, 21, 673–679. [DOI] [PubMed] [Google Scholar]
- Jaffe ES (2009) The 2008 WHO classification of lymphomas: implications for clinical practice and translational research. Hematology/The Education Program of the American Society of Hematology, 2009, 523–531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kane EV, Roman E, Becker N, Bernstein L, Boffetta P, Bracci PM, Cerhan JR, Chiu BC, Cocco P, Costas L, Foretova L, Holly EA, La Vecchia C, Matsuo K, Maynadie M, Sanjose S, Spinelli JJ, Staines A, Talamini R, Wang SS, Zhang Y, Zheng T & Kricker A (2012) Menstrual and reproductive factors, and hormonal contraception use: associations with non-Hodgkin lymphoma in a pooled analysis of InterLymph case-control studies. Annals of Oncology, 23, 2362–2374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nelson RA, Levine AM & Bernstein L (2001) Reproductive factors and risk of intermediate- or high-grade B-Cell non-Hodgkin’s lymphoma in women. Journal of Clinical Oncology, 19, 1381–1387. [DOI] [PubMed] [Google Scholar]
- Olsson H, Olsson ML & Ranstam J (1990) Late age at first full-term pregnancy as a risk factor for women with malignant lymphoma. Neoplasma, 37, 185–190. [PubMed] [Google Scholar]
- Prescott J, Lu Y, Chang ET, Sullivan-Halley J, Henderson KD, Clarke CA, Ma H, Temple-man C, Deapen D & Bernstein L (2009) Reproductive factors and non-Hodgkin lymphoma risk in the California Teachers Study. PLoS ONE, 4, e8135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wang SS, Luo J, Cozen W, Lu Y, Halley-Sullivan J, Voutsinas J, Zhong C, Song J, Lacey JV Jr, Weisenburger D & Bernstein L (2017) Sun sensitivity, indoor tanning and B-cell non-Hodgkin lymphoma risk among Caucasian women in Los Angeles County. British Journal of Haematology, 177, 153–156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhang Y, Holford TR, Leaderer B, Boyle P, Zahm SH, Zhang B, Zou K, Morton LM, Owens PH, Flynn S, Tallini G & Zheng T (2004) Menstrual and reproductive factors and risk of non-Hodgkin’s lymphoma among Connecticut women. American Journal of Epidemiology, 160, 766–773. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table SI. Distribution of B-cell NHL in Los Angeles County NHL case and control study (2004–2008)1