A population-based study on incidence trends of myeloma in the United States over 2000–2020

Myeloma is one of the most common types of haematological malignancies. We aimed to investigate the incidence rates of myeloma by sex, race, age, and histological subgroups in the United States (US) over 2000–2020. Data were retrieved from the the Surveillance, Epidemiology, and End Results (SEER) 22 database. The International Classification of Diseases for Oncology version 3 morphological codes 9731, 9732, and 9734 were assigned for solitary plasmacytoma of bone, plasma cell myeloma, and extraosseous plasmacytoma, respectively. Average annual percent change (AAPC) and the pairwise comparison with the parallelism and coincidence were reported. All estimates were reported as counts and age-adjusted incidence rates per 100,000 individuals. Over 2000–2019, most of myeloma cases were among those aged at least 55 years (85.51%), men (54.82%), and non-Hispanic Whites (66.67%). Among different subtypes, plasma cell myeloma with 193,530 cases had the highest frequency over the same period. Also, there was a significant decrease in the age-standardized incidence rate of myeloma across all races/ethnicities in both sexes within all age groups (AAPC: − 8.02; 95% confidence interval (CI): − 10.43 to − 5.61) and those aged < 55 (AAPC: − 8.64; 95% CI − 11.02 to − 6.25) from 2019 to November 2020. The overall trends of myeloma incidence rates were not parallel, nor identical. There was an increase in myeloma incidence in both sexes, with a highly increasing rate, particularly among younger Hispanic and non-Hispanic Black women over 2000–2019. However, a remarkable decline was observed in the incidence rates following the COVID-19 pandemic in 2020.


Statistical analysis
To estimate the delay age-standardized incidence rate (ASIR) of myeloma, the SEER 22 Research Limited-Field Data with Delay-Adjustment database for the period from 2000 to 2020 16 was collected from SEER*Stat, version 8.4.1.2 21.The purpose of modelling reporting delay is to update the current case count to account for expected future data revisions (including additions and deletions) 22 .These adjusted counts, as well as the associated delay model, are useful in detecting current cancer trends more precisely 22 .The case selection was limited to malignant myeloma cancer diagnosed with a known age.The delay model was then run using delay adjustment factors such as cancer site, registry, age group, race and ethnicity, and the year of diagnosis 23,24 .In addition, to estimate the ASIRs of myeloma subtypes, the SEER 22 Research Limited-Field Data database for the period from 2000 to 2020 17 was retrieved from SEER*Stat, version 8.4.1.2 21.The SEER*Stat version 8.4.1.2 21was used to estimate the ASIRs based on the 2000 US standard population and the accompanying 95% confidence intervals (CIs) with the Tiwari method 25 .
The Joinpoint Regression Program, version 5.0.2 26 , was employed to estimate the APCs, AAPCs 27 , joinpoint regression modelling, parallelism test, and coincident test 28 for ASIR 29 .Also, we used R software, version 4.3.2(Vienna, Austria) for some data visualization like the age plots.The first year of the COVID-19 pandemic was 2020, which had a significant influence on the health system, leading to reductions in cancer screening and diagnosis, and resulted in a decrease in the 2020 incidence rates for the majority of cancer sites.As a result, the 2020 incidence data may induce bias in cancer incidence estimations, hence it was omitted from Joinpoint trends and solely displayed in graphics 30 .The ASIR of myeloma APCs were calculated by producing the best fit of least-squares regression lines on the natural logarithm of the ASIR, with the year of diagnosis as a regressor variable.The minimal number of observations between two joinpoints and from the joinpoint to either end of the data was set to two.Model selection was accomplished using the weighted Bayesian Information Criteria technique 31 .The empirical quantile method was used to obtain the 95% CIs of AAPCs 32 .The pairwise comparison using the parallelism test was used to see if the trends of the two groups were similar over time 28 .In addition, a pairwise comparison with the coincidence test was performed to see whether the rates of the two groups were identical over time.
Older adults (ages ≥ 55).Over 2000-2019, a total of 175,185 cases of myeloma were diagnosed in those aged ≥ 55 years.Among older adults, plasma cell myeloma was the most common subtype (95.17%).Moreover, the majority of cases were males (54.53%) and NHWs (69.10%).Delayed ASIR per 100,000 population was 34.18 (95% CI 33.95 to 34.40) in men and 21.97 (95% CI 21.81 to 22.12) in women (Table 2).The incidence rate of myeloma in older adults increased over 2000-2019 in males and females with AAPC of 1.22 (95% CI 1.00 to 1.54) and 0.91 (95% CI 0.75 to 1.05), respectively (Table 2).Also, there was a steady increase in the incidence rate in all races, with NHB men having the highest AAPC (1.99%, 95% CI 1.24 to 2.66) (Table 2).The parallel and identical trends of myeloma among older adults are provided in Table S1 and Table S2.
Younger adults (ages < 55).A total of 29,687 cases of myeloma cancer were reported over 2000-2019 among adults aged less than 55 years in the US.Most of the cases were plasma cell myeloma (90.30%).The majority of www.nature.com/scientificreports/cases occurred in men (56.48%) and NHW (52.23%).The delayed ASIR per 100,000 population was 1.51 (95% CI 1.48 to 1.53) in men and 1.13 (95% CI 1.11 to 1.15) in women.Over 2000-2019, there was a higher increase in the ASIR of myeloma in women than men (AAPC: 2.15 vs. 1.68).There was an increase in incidence rates within all groups, with Hispanic women having the highest AAPC (2.27%; 95% CI 1.27 to 3.49) (Table 2).The trends of myeloma incidence over 2000-2019 among young adults in the US were not identical nor parallel (Table S1 and Table S2).
Older adults (ages ≥ 55).A total of 166,721 cases of plasma cell myeloma were reported between 2000 and 2019 in the US among those aged ≥ 55 years.The majority of the older adults were men (54.24%) and NHWs (68.86%). in women.Men had higher AAPC for plasma cell myeloma incidence rate among older adults than women (0.97 vs. 0.68) (Table 4).Over 2000-2019, all races/ethnicities had significant increases in the incidence rates and NHB men had the highest increase (AAPC: 62.49; 95% CI 61.42 to 63.56) (Table 4).
Younger adults (ages < 55).Over 2000-2019, a total of 26,809 cases of plasma cell myeloma were reported in the US among those aged < 55 years.The majority of them were men (55.45%) and NHWs (51.85%).The ASIR per 100,000 population was 1.32 (95% CI 1.30 to 1.34) for men and women.The AAPCs were higher in women than men (2.25 vs. 2.71) (Table 4).All groups experienced an increase between 2000 and 2019, with NHB women having the highest rate of increase (AAPC: 2.11; 95% CI 1.24 to 3.08) (Table 4).
Older adults (ages ≥ 55).Over 2000-2019, a total of 2087 cases of extraosseous plasmacytoma among older adults in the US were reported.The majority of them were men (61.76%) and NHWs (75.55%).The ASIR per 100,000 population was 0. ).The ASIR per 100,000 population were 0.12 (95% CI 0.12 to 0.13) for men and 0.06 (95% CI 0.06 to 0.07) for women.The AAPC did not have any significant changes by sex and race/ethnicity over the study period (Table 6).

Discussion
This study presented the trends in incidence rates of myeloma in the US across a two-decade interval.To the best of our knowledge, this is the first US population-based study that utilized a large data integrated in the SEER program, focusing on different subtypes of myeloma to perform a comprehensive analysis of the epidemiological trends of myeloma within 2000-2020.Overall, our findings demonstrated an increasing trend of myeloma incidence in both sexes within 2000-2019.Men constituted the majority of myeloma cases (54.82%) and a male predominance was observed in all three morphological subtypes.The incidence rates of myeloma were increasing at a relatively higher rate in older aged men compared to older aged women.However, in individuals younger than 55, the incidence rates increased at a higher pace in women.While myeloma had most commonly occurred in NHWs throughout the study period, NHBs exhibited the highest increase in incidence rate both in men and women among all races/ethnicities.However, racial and ethnic disparities were evident in the trends of myeloma incidence in old and young adult participants.Accordingly, NHBs had the highest AAPC in both www.nature.com/scientificreports/older men and women.However, in younger individuals, Hispanic women and NHB men exhibited the highest AAPC.Collectively, the concerning rise in the incidence rate of myeloma among younger women, specifically among young women of Hispanic ethnicity, should be carefully acknowledged when formulating public health strategies to address these worrisome patterns.Our analysis of plasma cell myeloma incidence trends, which is the most prevalent and extensively studied subtype of myeloma, revealed a consistent pattern similar to the overall trend observed for myeloma.However, NHBs exhibited the highest rate of increase in both younger and older adults across both sexes.In a recent study by Huang et al., an increasing trend of myeloma incidence was reported globally from 2001 to 2019 33 .According to their report and consistent with our observed trends, the incidence of myeloma in the US increased both in men and women, with a higher AAPC observed for men 33 .Previous reports on the incidence rate of myeloma between 1993 and 2012 in the US, have also pointed out the increased incidence and a tendency for younger age at diagnosis; however, the only significant increases were reported to occur in NHW women and men, as well as NHB men 34 .Our findings on the other hand, suggest an increase among all races/ethnicities with higher rates of increase in NHBs.Moreover, as observed in the general trend of myeloma, the highest AAPC for myeloma in our study also belongs to the younger women, suggesting that these individuals should be considered at risk with increasing rates of myeloma.www.nature.com/scientificreports/According to our findings, extraosseous plasmacytoma was the only subtype with decreasing AAPC in both men and women.Racial disparities were also observed in the incidence rate of extraosseous plasmacytoma, as NHBs demonstrated the highest rate of decrease in both sexes across all ages.However, in older ages, NHWs witnessed the highest rate of decline in incidence rate.On the other hand, no significant changes in the incidence rate of solitary plasmacytoma of bone were observed throughout the study period, except for NHW women who exhibited significant AAPC increase, which was pronounced in the older age women.While plasmacytoma and myeloma are plasma cell disorders that share certain cytological and immunophenotypic characteristics, their tendency to affect different sites suggest that variations exist between these entities 35,36 .A previous study on the incidence of myeloma subtypes in the US has highlighted that the incidence of myeloma was 16 times higher than plasmacytoma overall, and the incidence of plasmacytoma of the bone was 40% higher than extraosseous plasmacytoma 37 .Similar to our findings, rates for plasmacytomas were higher among males, but with increasing incidence over 1992-2004 37 .A more recent study on the US population reported overall ASIRs among adults to be 0.45 for solitary plasmacytoma and 0.12 for extraosseous plasmacytoma and the incidence rates were higher in men in both diseases 38 .
In order to address the impact of the COVID-19 pandemic on the reported incidence of myeloma in the US, we analyzed the corresponding data for 2020 and the change in AAPC compared to the preceding year.Findings revealed a remarkable decrease in the ASIR of myeloma within all races and both sexes following the outbreak of the pandemic.However, the differential impact of the pandemic was observed on different age groups and races.More precisely, the decline in ASIR was highly pronounced in older age groups, as no significant changes in the ASIR of myeloma were observed in the < 55 age group across any race or sex groups.Notably, male Hispanics were the only group who were not affected by the pandemic in any of the age groups.Overall, findings suggest that the incidence of myeloma among older individuals of all races except for male Hispanics was significantly affected by the COVID-19 pandemic.Supporting these observations, a recent umbrella review has identified a substantial decrease in screening and diagnosis of several cancers during the pandemic, which was more remarkable in regions that implemented a lockdown strategy 39 .Among these cancers, myeloma is no exception as previous global reports have indicated that the number of newly diagnosed myeloma cases was reduced in 2020 compared to 2019 11 .The reduced incidence of myeloma during the COVID-19 pandemic, as observed in our study and previous reports on other cancer entities 40,41 , could be attributed to impaired screening and utilization of diagnostic measures during the pandemic 42 .The overload of medical centers with COVID-19, as well as the reluctance of patients to seek diagnostic procedures might have resulted in underreporting of cancer patients.Supporting this, studies indicate that cancer biopsies, including bone marrow biopsies which are crucial in diagnosing hematological malignancies, have remarkably decreased with the outbreak of the pandemic 43 .
The underreporting and delayed diagnosis of myeloma has resulted in increased disease burden post-pandemic.Accordingly, a report by Carmichael et al. indicated an increasing presentation of myeloma cases to emergency services with heightened rates of osseous and extra-medullary manifestations 44 .Findings of another study have also confirmed decreased survival of myeloma cases after the pandemic 11 .Moreover, reports from the early COVID-19 pandemic have also highlighted that compared with the pre-pandemic era, myeloma patients during the COVID-19 pandemic were less likely to initiate treatment or had initiated treatment later 45 .Therefore, both delayed diagnosis and postponed treatment initiation might contribute to decreased survival.The delay in treatment and changes in treatment pattern, may also cause the pandemic to leave a mark on the survival of myeloma patients in the subsequent years 44 .Although the long-term effects of these changes may not be observable for a considerable period, it is evident that the COVID-19 pandemic, and its broader influence on healthcare delivery, have led to alterations in the epidemiology and burden of individuals with myeloma and these changes could potentially have significant implications for the patients' future prospects.
There are several strengths and limitations to this study.This is the first study to utilize the updated SEER database with robust statistical measures to provide a comprehensive US population-based study on incidence trends of myeloma.Besides analyzing the general trends of myeloma incidence, our study provided subgroup analyses on distinct pathological subtypes, including plasma cell myeloma and two less prevalent, yet critical subtypes, to better elucidate the incidence rate across various pathologic classes.Due to the unneglectable impact of the COVID-19 pandemic on cancer diagnosis, we also provided a detailed overview of changes in incidence rates of myeloma in the first year of the pandemic outbreak compared to the preceding year across all races, age groups and sexes.However, our findings might be influenced by the limitations of the SEER database.For instance, misclassification bias due to the collection of demographic data, including race and ethnicity, from various sources such as administrative databases, patient intake or provider notes is probable.Furthermore, there is uncertainty regarding whether self-identification of race and ethnicity accurately reflects ancestry or if it is primarily based on cultural factors, particularly in individuals who have mixed racial backgrounds.This ambiguity adds complexity to the interpretation of data related to race and ethnicity and should be taken into consideration when generalizing our findings.Additionally, our analysis was not sub-grouped according to race and ethnicity groups of American Indian/Alaskan Native, Asian/Pacific Islander, and Native Hawaiian due to the small sample size.Therefore, our findings may not fully represent the characteristics of individuals from these specific racial and ethnic backgrounds.The diagnostic methods of cancers can influence the incidence trends of myeloma and lead to under-or over-estimation of cancers.So, providing valid and reliable information for diagnostic methods and development of additional biomarkers data in SEER can improve the estimation of incidence of myeloma in future SEER data 46 .Also, there might be some other potential contributing factors to myeloma such as socioeconomic status that were not the aim of the study and can be considered in future studies.Moreover, data of some potential factors like occupation was not available in the current SEER iteration and can be taken into account for next SEER versions.Regarding the effects of COVID-19, there was a decreasing trend in the myeloma diagnosis by 14% in 2020 compared with 2019 at the global level 11 .The diagnosis of hematological malignancies was also decreased in the US during the COVID-19 pandemic compared with pre-pandemic period, especially due to the lockdown strategies 47 .As a result, the incidence data in 2020 was excluded from joinpoint trends.Nevertheless, it should be acknowledged as a study limitation and be considered in the interpretation of results.

Conclusions
There was an increase in myeloma incidence in both sexes, with a highly increasing rate, particularly among younger Hispanic and NHB women between 2000 and 2019.The findings of our study also underscore the significant impact of the COVID-19 pandemic on the reported incidence rate of myeloma in 2020 across most races/ethnicities, particularly in older ages.

Figure 1 .
Figure 1.Age-adjusted incidence rate of myeloma over 2000-2019 and in 2020 in the United States, by histologic type (A), sex (B), race/ethnicity (C), and age (D).APC annual percent change.*Represent p-value less than 0.05.

Table 1 .
Demographic characteristics of myeloma cases by the year of diagnosis.NHW Non-Hispanic White, NHB non-Hispanic Black.

Table 5 .
Counts and age-standardized rate of extraosseous plasmacytoma incidence per 100,000 and average annual percent change from 2000 to 2019 in the United States, by age, sex, and race.NHW Non-Hispanic White, NHB Non-Hispanic Black, ASIR Age-standardized incidence rate, CI Confidence interval, AAPC Average annual percent change.

Table 6 .
Counts and age-standardized rate of solitary plasmacytoma of bone incidence per 100,000 and average annual percent change from 2000 to 2019 in the United States, by age, sex, and race.NHW Non-Hispanic White, NHB Non-Hispanic Black, ASIR Age-standardized incidence rate, CI Confidence interval, AAPC Average annual percent change, NA Not available.