The association between renal cell carcinoma and multiple myeloma: insights from population-based data

Authors


Rohit P. Ojha, 44 Binney Street, SM 271, Boston, Massachusetts 02115, USA. e-mail: rohit_ojha@dfci.harvard.edu

Abstract

Study Type – Prevalence (population based cohort)

Level of Evidence 3b

What’s known on the subject? and What does the study add?

Several case-series have hypothesized a potential association between renal cell carcinoma and multiple myeloma. Nonetheless, this hypothesis has not been systematically explored in a population-based setting with sufficient sample size to estimate a magnitude of association.

Our analyses revealed a bidirectional relation between renal cell carcinoma and multiple myeloma, which typically indicates that common risk factors influence both malignancies. Our findings may be useful for raising awareness among clinicians that a diagnosis of multiple myeloma may be within the spectrum of second malignancies among patients with renal cell carcinoma and that a diagnosis of renal cell carcinoma may be within the spectrum of second malignancies among patients with multiple myeloma.

OBJECTIVE

• To evaluate the hypothesis of an association between renal cell carcinoma and multiple myeloma.

PATIENTS AND METHODS

• Data from nine population-based registries in the Surveillance, Epidemiology and End Results programme were used to evaluate two separate cohorts of patients diagnosed between 1973 and 2006: patients diagnosed with renal cell carcinoma as a primary malignancy (n= 57 190) and patients diagnosed with multiple myeloma as a primary malignancy (n= 34 156).

• We estimated standardized incidence ratios (SIRs) with corresponding 95% confidence intervals (CIs) by dividing the number of observed cases of multiple myeloma within the renal cell carcinoma cohort and the number of renal cell carcinoma cases within the multiple myeloma cohort by the number of expected cases for each malignancy in the US general population.

RESULTS

• The renal cell carcinoma cohort yielded 88 multiple myeloma cases during 293 511 person-years of follow up. Patients with renal cell carcinoma had a higher relative risk of multiple myeloma than the general population (SIR = 1.51, 95% CI 1.21–1.85).

• The multiple myeloma cohort yielded 69 renal cell carcinoma cases during 100 804 person-years of follow up. Patients with multiple myeloma had a higher relative risk of renal cell carcinoma than the general population (SIR = 1.89, 95% CI 1.47–2.40).

CONCLUSION

• Our analyses revealed a bidirectional association between renal cell carcinoma and multiple myeloma, which typically indicates shared risk factors.

Abbreviations
SEER

Surveillance, Epidemiology and End Results

SIR

standardized incidence ratios.

INTRODUCTION

Survivors of a primary malignancy have an increased risk of developing a second primary malignancy because of shared risk factors (genetic and environmental), antecedent cancer therapy, or other factors [1]. Patients with renal cell carcinoma are particularly susceptible to second primary malignancies [2,3]; previous studies have identified associations between renal cell carcinoma and cancers of the bladder [2–5], prostate [2–4,6], colon [2–4] and lung [2,4,5] as well as malignant melanoma [2,7,8] and non-Hodgkin lymphoma [2,4,8–10]. Recently, several case-series analyses have prompted investigators to postulate an association between renal cell carcinoma and multiple myeloma [10–16]. Although case-series are vital for generating hypotheses, this type of study design is insufficient for determining an association because of the small sample size and the lack of a comparison group [17]. A study with sufficient sample size and an appropriate comparison group is necessary to determine if an association exists between renal cell carcinoma and multiple myeloma. We used population-based data from the Surveillance, Epidemiology and End Results (SEER) registries to evaluate the hypothesis of an association between renal cell carcinoma and multiple myeloma and so provide insight regarding the necessity for clinicians to be aware of the potential for a diagnosis of a second primary multiple myeloma among patients with renal cell carcinoma.

PATIENTS AND METHODS

We used data from nine population-based registries in the SEER programme [18] to evaluate two separate patient cohorts diagnosed with a primary malignancy between 1 January 1973 and 31 December 2006: patients diagnosed with renal cell carcinoma and patients diagnosed with multiple myeloma. We compiled person-years of observation for individuals in each cohort after initial primary cancer diagnosis until date of death, the end of the observation period (31 December 2006), or subsequent primary diagnosis of multiple myeloma in all patients with an initial diagnosis of renal cell carcinoma or subsequent primary diagnosis of renal cell cancer in all patients with an initial diagnosis of multiple myeloma.

We estimated overall, age-specific (<50 years, 50–59 years, 60–69 years, 70–79 years and ≥80 years), and gender-specific standardized incidence ratios (SIRs) with corresponding 95% CIs for multiple myeloma incidence in the renal cell carcinoma cohort and for renal cell carcinoma incidence in the multiple myeloma cohort using the standard formula (SIR = observed number of cases/expected number of cases) [5,19]. We also estimated overall and gender-specific SIRs with corresponding 95% CIs for defined periods of follow up after diagnosis (<1 year, 1–5 years, 5–10 years, >10 years) in each cohort. Briefly, the numerator (observed number of cases) of the SIR was determined by enumerating subsequent multiple myeloma cases observed for patients with renal cell carcinoma and subsequent renal cell carcinoma cases observed for patients with multiple myeloma. The denominator (expected number of cases) of the SIR was determined by multiplying the SEER-defined standardized incidence rate (age; gender; ethnicity, i.e. Black, White, Other; and calendar-year adjusted incidence rate observed in the general US population) of each cohort by the accumulated person-years within each cohort to obtain the expected number of cases of multiple myeloma for the renal cell carcinoma cohort and of renal cell carcinoma for the multiple myeloma cohort. All analyses were performed using the multiple primary-standardized incidence ratio session of SEER*Stat (National Cancer Institute, Bethesda, MD, USA) [20].

RESULTS

Overall, 57 190 patients (mean age = 61.2 years; sd= 16.4; 62.2% men) were diagnosed with renal cell carcinoma between 1973 and 2006. This cohort yielded 88 multiple myeloma cases during 293 511 person-years of follow up (median = 2.50 years, interquartile range 0.58–7.42). Patients with renal cell carcinoma had a higher overall relative risk of multiple myeloma than the general population (SIR = 1.51, 95% CI 1.21–1.85). Age-specific analyses revealed that the highest relative risk of multiple myeloma after renal cell carcinoma diagnosis was observed among patients aged 50–59 years (50–59 years: SIR = 3.19, 95% CI 1.83–5.19; <50 years: SIR = 1.33, 95% CI 0.03–7.42; 60–69 years: SIR = 1.07, 95% CI 0.61–1.74; 70–79 years: SIR = 1.20, 95% CI 0.79, 1.73; >80 years: SIR = 1.88, 95% CI 1.24–2.73). Table 1 summarizes the overall and age-specific relative risks of multiple myeloma incidence among patients with renal cell carcinoma.

Table 1.  Age-specific relative risks of (A) multiple myeloma incidence among patients with renal cell carcinoma and (B) renal cell carcinoma incidence among patients with multiple myeloma compared with the US population, 1973–2006
(A) Multiple myeloma incidence among patients with renal cell carcinoma
Age groupIndividuals with RCCPerson-years at risk of MMObserved MM casesExpected MM casesSIR (95% CI)
Overall57 190 293 5118858.461.51 (1.21–1.85)
<50 years10 86056 100 10.751.33 (0.03–7.42)
50–59 years16 95957 861165.013.19 (1.83–5.19)
60–69 years22 19376 2321614.901.07 (0.61–1.74)
70–79 years20 45868 4742823.401.20 (0.79–1.73)
>80 years 11 20034 8442714.391.88 (1.24–2.73)
(B) Renal cell carcinoma incidence among patients with multiple myeloma
Age groupIndividuals with MMPerson-years at risk of RCCObserved RCC casesExpected RCC casesSIR (95% CI)
  1. RCC, renal cell carcinoma; MM, multiple myeloma; SIR, standardized incidence ratio; 95% CI, 95% confidence interval.

Overall34 156100 8046936.451.89 (1.47–2.40)
<50 years 2 720  7 954 2 0.603.32 (0.40–12.01)
50–59 years 6 84518 088 11 3.922.81 (1.40–5.03)
60–69 years 11 68328 7961910.581.80 (1.08–2.80)
70–79 years13 61630 2072614.481.80 (1.17–2.63)
>80 years 8 63615 75911 6.871.60 (0.80–2.86)

The highest relative risk of multiple myeloma was observed within the first year after renal cell carcinoma diagnosis (<1 year: SIR = 3.82, 95% CI 2.59–5.42; 1–5 years: SIR = 1.07, 95% CI 0.67–1.62; 5–10 years: SIR = 1.40, 95% CI 0.86–2.13; >10 years: SIR = 0.95, 95% CI 0.52–1.59). Thirty-five percent (31/88) of all incident multiple myeloma cases were diagnosed within the first year after renal cell carcinoma diagnosis. The relative risk of multiple myeloma was higher among men and women with renal cell carcinoma than men and women in the general population, respectively (men: SIR = 1.35, 95% CI 1.02–1.76; women: SIR = 1.84, 95% CI 1.27–2.56). The highest relative risk of multiple myeloma among men was observed within the first year after renal cell carcinoma diagnosis (<1 year: SIR = 2.33, 95% CI 1.24–3.98; 1–5 years: SIR = 1.06, 95% CI 0.59–1.75; 5–10 years: SIR = 1.55, 95% CI 0.89–2.52; >10 years: SIR = 1.01, 95% CI 0.48, 1.85). Similarly, the highest relative risk of multiple myeloma among women was observed within the first year after renal cell carcinoma diagnosis (<1 year: SIR = 7.11, 95% CI 4.22–11.24; 1–5 years: SIR = 1.09, 95% CI 0.44–2.25; 5–10 years: SIR = 1.05, 95% CI 0.34–2.45; >10 years: SIR = 0.83, 95%CI 0.23–2.13). Table 2 summarizes the overall and gender-specific relative risks of multiple myeloma incidence among patients with renal cell carcinoma by duration of follow up since diagnosis of renal cell carcinoma.

Table 2.  Relative risks of multiple myeloma incidence among patients with renal cell carcinoma compared with the US population, 1973–2006
GroupLatency (years)Individuals with RCCPerson-years at risk of MMObserved MM casesExpected MM casesSIR (95% CI)
  1. RCC, renal cell carcinoma; MM, multiple myeloma; SIR, standardized incidence ratio; 95% CI, 95% confidence interval.

Overall 57 190 293 5118858.461.51 (1.21–1.85)
 <157 19044 940318.123.82 (2.59–5.42)
1–538 574109 9892220.561.07 (0.67–1.62)
5–1020 16673 3232115.051.40 (0.86–2.13)
>1010 35865 2591414.730.95 (0.52–1.59)
Men 35 552176 1415439.931.35 (1.02–1.76)
 <135 55227 931135.592.33 (1.24–3.98)
1–523 93167 4791514.131.06 (0.59–1.75)
5–1012 21543 8171610.291.55 (0.89–2.52)
>10 6 11436 914109.921.01 (0.48–1.85)
Women 21 638 117 3703418.531.84 (1.27–2.56)
 <121 63817 009182.537.11 (4.22–11.24)
1–514 64342 51076.421.09 (0.44–2.25)
5–107 95129 50654.761.05 (0.34–2.45)
>104 24428 34544.810.83 (0.23–2.13)

For the cohort of patients with multiple myeloma, 34 156 patients (mean age = 67.9 years; sd 12.3; 52.6% men) were diagnosed with multiple myeloma between 1973 and 2006. This cohort yielded 69 renal cell carcinoma cases during 100 804 person- years of follow up (median = 1.83 years, interquartile range 0.58–4.00). Patients with multiple myeloma had a higher overall relative risk of renal cell carcinoma than the general population (SIR = 1.89, 95% CI 1.47–2.40). Age-specific analyses revealed that the highest relative risk of renal cell carcinoma after multiple myeloma diagnosis was observed among patients aged <50 years (<50 years: SIR = 3.32, 95% CI 0.40–12.01; 50–59 years: SIR = 2.81, 95% CI 1.40–5.03; 60–69 years: SIR = 1.80, 95% CI 1.08–2.80; 70–79 years: SIR = 1.80, 95% CI 1.17–2.63; >80 years: SIR = 1.60, 95% CI 0.80–2.86). Table 1 summarizes the overall and age-specific relative risks of renal cell carcinoma incidence among patients with multiple myeloma.

The highest relative risk of renal cell carcinoma was observed within the first year after multiple myeloma diagnosis (<1 year: SIR = 4.14, 95% CI 2.93–5.69; 1–5 years: SIR = 1.15, 95% CI 0.71–1.76; 5–10 years: SIR = 1.27, 95% CI 0.55–1.49; >10 years: SIR = 0.73, 95% CI 0.09–2.64). Fifty-five percent (38/69) of all incident renal cell carcinoma cases were diagnosed within the first year after multiple myeloma diagnosis. The relative risk of renal cell carcinoma was higher among men and women with multiple myeloma than men and women in the general population (men: SIR = 2.03, 95% CI 1.51–2.66; women: SIR = 1.60, 95% CI 0.95–2.52). The highest relative risk of renal cell carcinoma among men was observed within the first year after multiple myeloma diagnosis (<1 year: SIR = 4.26, 95% CI 2.81–6.20; 1–5 years: SIR = 1.27, 95% CI 0.73–2.06; 5–10 years: SIR = 1.37, 95% CI 0.50–2.97; >10 years: SIR = 1.08, 95% CI 0.13–3.89). Similarly, the highest relative risk of renal cell carcinoma among women was observed within the first year after multiple myeloma diagnosis (<1 year: SIR = 3.88, 95% CI 1.94–6.95; 1–5 years: SIR = 0.89, 95% CI 0.29–2.07; 5–10 years: SIR = 1.04, 95% CI 0.13–3.75; >10 years: SIR = 0.00, 95% CI 0.00–4.20 – no renal cell carcinoma cases among women during this interval). Table 3 summarizes the overall and gender-specific relative risks of renal cell carcinoma incidence among patients with multiple myeloma by duration of follow up since diagnosis of multiple myeloma. Furthermore, Figure 1 illustrates the overall trend in relative risks of multiple myeloma incidence among patients with renal cell carcinoma and of renal cell carcinoma among patients with multiple myeloma by duration of follow up since diagnosis of the primary malignancy.

Table 3.  Relative risks of renal cell carcinoma incidence among patients with multiple myeloma compared with the US population, 1973–2006
GroupLatency (years)Individuals with MMPerson-years at risk of RCCObserved RCC casesExpected RCC casesSIR (95% CI)
  1. RCC, renal cell carcinoma; MM, multiple myeloma; SIR, standardized incidence ratio; 95% CI, 95% confidence interval.

Overall 34 156100 8046936.451.89 (1.47–2.40)
 <134 15626 568389.174.14 (2.93–5.69)
1–522 52251 0462118.231.15 (0.71–1.76)
5–106 44816 66986.321.27 (0.55–2.49)
>101 5836 52222.740.73 (0.09–2.64)
Men 17 97753 0485125.182.03 (1.51–2.66)
 <117 97713 999276.344.26 (2.81–6.20)
1–511 88926 9231612.591.27 (0.73–2.06)
5–103 3838 83364.391.37 (0.50–2.97)
>10 8333 29321.861.08 (0.13–3.89)
Women 16 17947 75618 11.271.60 (0.95–2.52)
 <116 17912 568 112.833.88 (1.94–6.95)
1–510 63324 12255.640.89 (0.29–2.07)
5–103 0657 83721.931.04 (0.13–3.75)
>10 7503 22900.880.00 (0.00–4.20)
Figure 1.

Relative risks of multiple myeloma incidence among patients with renal cell carcinoma and renal cell carcinoma incidence among patients with multiple myeloma by duration since diagnosis of primary malignancy.

DISCUSSION

This population-based investigation was designed to evaluate the hypothesis generated from several recent case-series that renal cell carcinoma and multiple myeloma are associated. Our analyses revealed a bidirectional association between renal cell carcinoma and multiple myeloma; the relative risk of multiple myeloma incidence was 51% higher among patients with renal cell carcinoma than in the general population and the relative risk of renal cell carcinoma incidence was 89% higher among multiple myeloma patients than in the general population. The highest overall relative risks were observed within the first year after diagnosis of the primary malignancy. In general, the bidirectional association between these malignancies suggests shared risk factors rather than treatment-related consequences [21], which is consistent with other reported second primary malignancies among patients with renal cell carcinoma [2]. Furthermore, chemotherapy and radiation therapy are not conventional treatment options for renal cell carcinoma, which decreases the potential for treatment-related concerns [2].

Renal cell carcinoma and multiple myeloma are more frequently diagnosed among older adults and there are steady increases in age-specific incidence rates for both malignancies [22]. Therefore, we anticipated gradually higher relative risks of second primary multiple myeloma and renal cell carcinoma with increasing age. However, the age-specific relative risk estimates of multiple myeloma incidence after renal cell carcinoma indicate a bimodal risk distribution with patients aged 50–59 years and >80 years at highest risk, whereas the age-specific relative risk estimates of renal cell carcinoma incidence after multiple myeloma are relatively consistent across the age groups examined in our study. Consequently, age alone may not be a sufficient explanation for the bidirectional association between these malignancies.

Our results also do not indicate a notable influence of gender on the relative risk of second primary multiple myeloma or renal cell carcinoma. Although the gender-specific relative risk estimates of both malignancies are modestly disparate, the overlapping confidence intervals suggest that notable differences between genders are not apparent after accounting for random variation. However, women have a remarkably higher relative risk of multiple myeloma incidence within 1 year after renal cell carcinoma diagnosis compared with men. The lack of evidence regarding an effect of reproductive or hormonal factors on multiple myeloma incidence [23] could indicate that the observed higher relative risk of multiple myeloma incidence may be related to disparate surveillance patterns between genders, but this gender disparity was not apparent when renal cell carcinoma followed multiple myeloma. Therefore, factors other than medical surveillance may account for this initially higher relative risk of myeloma incidence among women with renal cell carcinoma.

Although not necessarily specific to this association, certain genetic risk factors may be shared between renal cell carcinoma and multiple myeloma that contribute to the observed bidirectional association. For example, c-met oncogene mutations are well-recognized in hereditary papillary renal cell carcinoma [24,25]. Hereditary papillary renal cell carcinoma is most often observed in older age [25], an attribute that is consistent with multiple myeloma incidence. More importantly, c-met expression was recently implicated in myeloma cell proliferation through feedback loops with interleukin-6, an inflammatory cytokine that has a critical role in the development and growth of myeloma cells [26]. Therefore, c-met expression may be a candidate for elucidating the association between renal cell carcinoma and multiple myeloma.

Lifestyle-related risk factors such as obesity may also contribute to the observed association. Obesity is a strong risk factor for renal cell carcinoma [24,25] and has modest but consistent effects on multiple myeloma incidence [23,27]. Obesity also offers a plausible alternative for the hypothesis proposed in previous studies [10,11,13] that increased interleukin-6 expression in renal cell carcinoma with bone metastasis may have a role in the development of multiple myeloma. Foremost, we observed only two cases of multiple myeloma associated with metastatic renal cell carcinoma throughout the duration of our study period (data not shown). Furthermore, obesity is characterized by marked changes in adipose tissue, such as increased number and size of adipocytes [28]. Adipose tissue is a major source of inflammatory mediators, particularly interleukin-6 [29,30]. Therefore, obesity may facilitate the creation of a microenvironment that supports the development of renal cell carcinoma and multiple myeloma.

Occupational exposures offer a seemingly appealing explanation for the higher relative risks of renal cell carcinoma and multiple myeloma among men. Excess relative risks of renal cell carcinoma after occupational exposure to certain chemicals have been observed only among men [25] but the evidence regarding the effect of occupational exposures on multiple myeloma incidence is inconsistent [23]. Methodological limitations in studies of occupational exposures are problematic for interpreting potential effects, particularly for uncommon malignancies such as multiple myeloma [23]. Furthermore, the elucidation of specific chemical exposures is challenging when certain occupations are subject to multiple chemical exposures. Therefore, the potential remains that certain occupational exposures are currently unrecognized risk factors for renal cell carcinoma and multiple myeloma.

Our results should be considered in the context of certain biases. Although later intervals of follow up since diagnosis of the primary malignancy exhibit modest increases in overall relative risks of renal cell carcinoma and multiple myeloma, the highest overall and gender-specific relative risks of both malignancies were observed within the first year of diagnosis, which suggests incidental diagnoses. Renal cell carcinoma diagnosis is frequently incidental because of advances in imaging [31], but both patients with renal cell carcinoma and those with multiple myeloma undergo comprehensive diagnostic procedures and clinical surveillance whereas individuals in the general population, who constituted the comparison group for estimating relative risks, are less likely to receive medical attention if healthy. Therefore, detection bias because of systematic differences in medical surveillance between patients with renal cell carcinoma or multiple myeloma and the general population may partially explain the notable initial increases in relative risks. However, lead-time bias consequent to medical surveillance, which is typically considered in the context of cancer screening and mortality [32], may have a deceptive role in our analysis by producing initially high relative risk estimates that could be misinterpreted for a lack of sustained association between renal cell carcinoma and multiple myeloma.

In summary, our population-based investigation was designed to evaluate the hypothesis generated from several recent case-series that renal cell carcinoma and multiple myeloma are associated. Our analyses revealed a bidirectional association between renal cell carcinoma and multiple myeloma, which suggests shared risk factors rather than treatment-related consequences [21]. The apparent potential for shared risk factors between these malignancies may be useful evidence for generating new hypotheses or prompting re-evaluations of potentially abandoned hypotheses regarding risk factors for renal cell carcinoma and particularly for multiple myeloma because few risk factors have been identified. Furthermore, our findings may be useful for raising awareness among clinicians that a diagnosis of multiple myeloma may be within the spectrum of second primary malignancies among patients with renal cell carcinoma (along with those reported in previous studies [2–10]) and that a diagnosis of renal cell carcinoma may be within the spectrum of second primary malignancies among patients with multiple myeloma.

ACKNOWLEDGEMENTS

This study was assigned exempt status by the University of North Texas Health Science Center Institutional Review Board.

CONFLICT OF INTEREST

None declared.

Ancillary