
                                       
      Supplemental Information for the Beryllium Health Effects Section 


Table A.1 -- Summary of Beryllium Lung Cancer Epidemiological Studies


(%) Prevalence




Reference
Study type
Sensitization
CBD
Range of exposure
measurements
Exposure-response
relationship
Study limitations
Additional
comments
Studies Conducted Prior to BeLPT
 
 
 
 
 
 

Hardy and Tabershaw, 1946 (Document ID 1516)
Case-series
N/A
N/A
N/A

Selection bias
Small sample size.
Hardy, 1980 (Document ID 1514)
Case-series
N/A
N/A
N/A

Selection bias
Small sample size.
Machle et al., 1948 (Document ID 1461)
Case-series
N/A

Semi-quantitative
Yes
Selection bias
Small sample size; unreliable exposure data.
Eisenbud et al., 1949 (Document ID 1949)
Case-series
N/A

Average concentration: 350-750 ft from plant -- 0.05-0.15 μg/m[3]; <350 ft from plant -- 2.1 μg/m[3] 


Non-occupational; ambient air sampling.
Lieben and Metzner, 1959 (Document ID 1343)

N/A
N/A
No quantitative exposure data


Family member contact with contaminated clothes.
Hardy et al., 1967 (Document ID 1515)
Case Registry Review
N/A
N/A
Incomplete exposure concentration data



Hasan and Kazemi, 1974 (Document ID 0451)







Eisenbud and Lisson, 1983 (Document ID 1296)



1-10



Stoeckle et al., 1969 (Document ID 0447)
Case-series (60 cases)



No
Selection bias
Provided information regarding progression and identifying sarcoidosis from CBD.
Studies Conducted Following the Development of the BeLPT
 
 
 
 
 
 

Beryllium Mining and Extraction 
 
 
 
 
 
 

Deubner et al., 2001 (Document ID 1543)
Cross-sectional 
(75 workers)
4.0 (3 cases)
1.3 (1 case)
Mining, milling -- range 0.05-0.8 μg/m[3]; Annual maximum 0.04-165.7 μg/m[3] 
No
Small sample size
Personal sampling.

Beryllium Metal Processing and Alloy Production 
 
 
 
 
 
 
 

Kreiss et al., 1997 Document ID 1360)



Cross-sectional study of 627 workers


6.9 (43 cases)
4.6 (29 cases)
Median -- 1.4 μg/m[3] 
No
Inconsistent BeLPT results between labs
Short-term Breathing Zone sampling.

Schuler et al., 2012 (Document ID 0473)
Cross-sectional study of 264 workers

9.8 (26 cases)
2.3 (6 cases)
Mean annual 0.88 μg/m[3]; Mean cumulative 2.11 μg/m[3]; Mean'highest job' 1.66 μg/m[3] 


Study was able to establish a significant increasing trend between CBD and cumulative exposure and able to establish a significant increasing trend between BeS and average TWA exposure as well as' highest job' exposure
Bailey et al., 2010 (Document ID 0610)
Longitudinal study of 290  program workers (206 partial program, 84 full program). Compari- son group of 258 pre-program workers 
2.1 (6 cases)
Not reported


Study reported prevalence rates for pre enhanced control-program, partial enhanced control program, and full enhanced control program

Thomas et al., 2009 (Document ID 1061)
Longitud- inal study of 82 workers. Compari- son group of 43 legacy workers 
3.8 (3 cases)
0.0 
Used exposure profile from Schuler study

Authors noted workers may have been sensitized prior to available screening, underestimating sensitization rate in legacy workers
Instituted PPE to reduce dermal exposures.
Rosenman et al., 2005 (Document ID 1352)
Cross-sectional study of 577 workers
10.4 (60 cases)
7.6 (44 cases)
Mean average range -- 7.1-8.7 μg/m[3];Mean peak range -- 53-87 μg/m[3];Mean cumulative range -- 100-209 μg/m[3] 
No

Daily weighted average:  High exposures compared to other studies.

Beryllium Machining Operations 
 
 
 
 
 
 
 

Newman et al., 2001
(Document ID 1354)
Longitud- inal study of 235 workers
9.4 (22 cases)
6.4 (15 cases)

No

Engineering and administrative controls primarily used to control exposures.

Kelleher et al., 2001(Document ID 1363)
Case-control study of 20 cases and 206 controls
11.5 (machinists)2.9 (non-machinists)
11.5 (machinists)2.9 (non-machinists)
0.08-0.6 μg/m[3] -- lifetime weighted exposures
Yes

Identified 20 workers with Sensitization or CBD.

Madl et al., 2007 (Document ID 1056)
Longitud- inal study of 27 cases


Machining 1980-1995 median −0.33 μg/m[3]; 1996-1999 median -- 0.16 μg/m[3]; 2000-2005 median -- 0.09 μg/m[3]; Non-machining 1980-1995 median -- 0.12 μg/m[3]; 1996-1999 median -- 0.08 μg/m[3]; 2000-2005 median -- 0.06 μg/m[3] 
Yes

Personal sampling: Required evidence of granulomas for CBD diagnosis.

Beryllium Oxide Ceramics 
 
 
 
 
 
 

Kreiss et al., 1993 (Document ID 1478)
Cross-sectional survey of 505 workers
3.6 (18 cases)
1.8 (9 cases)

No


Kreiss et al., 1996 (Document ID 1477)
Cross-sectional survey of 136 workers
5.9 (8 cases)
4.4 (6 cases)
Machining median -- 0.6 μg/m[3]; Other Areas median -- <0.3 μg/m[3];
No
Small study population
Breathing Zone Sampling.
Henneberger et al., 2001 (Document ID 1313)
Cross-sectional survey of 151 workers
9.9 (15 cases)
5.3 (8 cases)
6.4% samples >2 μg/m[3]; 2.4% samples >5 μg/m[3];0.3% samples >25 μg/m[3] 
Yes
Small study population
Breathing zone sampling.
Cummings et al., 2007 (Document ID 1369)
Longitud- inal study of 93 workers

1.1 (1 case)
Production 1994-1999 median -- 0.1μg/m[3]; 2000-2003 median -- 0.04μg/m[3]; Administrative 1994-1999 median <0.2 μg/m[3]; 2000-2003 median -- 0.02 μg/m[3] 

Small sample size
Personal sampling was effective in reducing rates of new cases of sensitization.

Copper-Beryllium Alloy Processing and Distribution 
 
 
 
 
 
 

Schuler et al., 2005 (Document ID 0919)
Cross-sectional survey of 153 workers
7.0 (10 cases)
4.0 (6 cases)
Rod and Wire Production median -- 0.12 μg/m[3]; Strip Metal Production median -- 0.02 μg/m[3]; Production Support median -- 0.02 μg/m[3]; Administration 
median -- 0.02 μg/m[3] 

Small study population
Personal sampling.








Stanton et al., 2006 (Document ID 1070)
Cross-sectional study of 88 workers
1.1 (1 case)
1.1 (1 case)
Bulk Products Production median 0.04 μg/m[3]; Strip Metal Production median -- 0.03 μg/m[3]; Production support median -- 0.01 μg/m[3]; Administration median 0.01 μg/m[3] 

Study did not report use of PPE or respirators
Personal sampling.
Nuclear Weapons Production Facilities and Cleanup of Former Facilities 
 
 
 
 
 
 

Kreiss et al., 1989 (Document ID 1480)
Cross-sectional survey of 51 workers
11.8 (6 cases)
7.8 (4 cases)

No
Small study population

Kreiss et al., 1993 (Document ID 1479)
Cross-sectional survey of 895 workers
1.9 (18 cases)
1.7 (15 cases)

No
Study population includes some workers with no reported Be exposure

Stange et al., 1996 (0206)
Longitud- inal Study of 4,397 BHSP participants
2.4 (105 cases)
0.7 (29 cases)
Annual mean concentration 1970-1988 0.016 μg/m[3]; 1984-1987 1.04 μg/m[3] 
No

Personal sampling.
Stange et al., 2001 (Document ID 1403)
Longitud- inal study of 5,173 workers
4.5 (235 cases)
1.6 (81 cases)
No quantitative information presented in study
No

Personal sampling.
Viet et al., 2000 (Document ID 1344)
Case-control
74 workers sensitized
50 workers CBD
Mean exposure range: 0.083-0.622 μg/m[3] Maximum exposures: 0.54-36.8 μg/m[3] 
Yes
Likely underestimated exposures
Fixed airhead sampling away from breathing zone: Matched controls for age, sex, smoking.
Johnson et al (2001) (Document ID 1505)
194 workers
No screening for sensitization
1 case of CBD out of 194 workers
Authors estimated airborne beryllium concentrations did not exceed 2.0 ug/m[3]

Limited information on collection times limits the reliability of exposure data
No screening for sensitization
Arjomandi et al (2010) (Document ID 1275)
Cross sectional of 1875 workers
59 workers out of 1875 (3.1 percent)
5 workers out 1875 (0.3 percent)



10.2 percent CBD rate among sensitized workers (5/49)
Aluminum Smelting







Taiwo et al. (2010)
Cross sectional study of 1932 workers 
2  of 1932 with CBD

0.01-13.0 ug/m3 TWA, mean of 0.25 ug/m[3]


Only 60 percent of workers participate in study
Nilsen et al. (2010)
Case control (362 workers)


0.1 ug/m[3]  -  0.31 ug/m[3]






Table A.2 -- Summary of Mechanistic Animal Studies for Sensitization and CBD
Reference
Species
Study length
Dose or exposure concentration
Type of beryllium
Study results
Other information
Intratracheal (intrabroncheal) or Nasal Instillation
 
 
 
 
 
 
Barna et al., 1981
(Document ID1500) 
Guinea pig
3 month
10 mg-5μm particle size
beryllium oxide
Granulomas, interstitial infiltrate with fibrosis with thickening of alveolar septae
 
Barna et al., 1984
(Document ID 1301 and 1302))
Guinea pig
3 month
5 mg
beryllium oxide
Granulomatous lesions in strain 2 but not strain 13 indicating a genetic component
 
Benson et al., 2000 (Document ID 1498)
Mouse

0, 12.5, 25, 100μg; 0, 2, 8 μg
beryllium copper alloy; beryllium metal
Acute pulmonary toxicity associated with beryllium/copper alloy but not beryllium metal
 
Haley et al., 1994 (Document ID 1364)
Cynomolgus monkey
14, 60, 90 days
0, 1, 50, 150 μg0, 2.5, 12.5, 37.5 μg
Beryllium metal, beryllium oxide
Beryllium oxide particles were less toxic than the beryllium metal
 
Huang et al., 1992 (Document ID 1492)
Mouse

5 μg1-5 μg
Beryllium sulfate immunization; beryllium metal challenge
Granulomas produced in A/J strain but not BALB/c or C57BL/6

Votto et al., 1987 (Document ID 1305)
Rat
3 month
2.4 mg 8 mg/ml
Beryllium sulfate immunization; beryllium sulfate challenge
Granulomas, however, no correlation between T-cell subsets in lung and BAL fluid
 
Inhalation -- Single Exposure
 
 
 
 
 
 
Haley et al., 1989 and 1991 (Document ID 1366 and 1315)
Beagle dog
Chronic -- one dose
0, 6 μg/kg, 18 μg/kg
500 °C; 1000 °C beryllium oxide
Positive BeLPT results -- developed granulomas; low-calcined beryllium oxide more toxic than high-calcined
Granulomas resolved with time, no full-blown CBD.


Chronic -- one dose/2 year recovery
0, 17 μg/kg, 50 μg/kg
500 °C; 1000 °C beryllium oxide
Granulomas, sensitization, low-fired more toxic than high fired
Granulomas resolved over time.
Robinson et al., 1968 (1487)
Dog
Chronic
0. 115mg/m[3] 
Beryllium oxide, beryllium fluoride, beryllium chloride
Foreign body reaction in lung
 
Sendelbach et al., 1989 (Document ID 1289)
Rat
2 week
0, 4.05 μg/L
Beryllium as beryllium sulfate
Interstitial pneumonitis
 
Sendelbach and Witschi, 1987 (Document ID 1307)
Rat
2 week
0, 3.3, 7 μg/L
Beryllium as beryllium sulfate
Enzyme changes in BAL fluid
 
Inhalation -- Repeat Exposure
 
 
 
 
 
 
Conradi et al., 1971 (Document ID 1319)
Beagle dog
Chronic -- 2 year
0. 3300 μg/m[3], 4380 μg/m[3] once/month for 3 months
1400 °C beryllium oxide
No changes detected
May have been due to short exposure time followed by long recovery.

Macaca irus Monkey
Chronic -- 2 year
0. 3300 μg/m[3], 4380 μg/m[3] once/month for 3 months
1400 °C beryllium oxide
No changes detected
May have been due to short exposure time followed by long recovery.
Haley et al., 1992 (Document ID 1234)
Beagle dog
Chronic -- repeat dose (2.5 year intervals)
17, 50 μg/kg
500 °C; 1000 °C beryllium oxide
Granulomatous pneumonitis
 
Harmsen et al., 1986 (Document ID 1257)
Beagle dog5 dogs per group
Chronic
0, 20 μg/kg, 50 μg/kg
500°C; 1000 °C beryllium oxide
 
 
Dermal or Intradermal
 
 
 
 
 
 
Kang et al., 1977 (Document ID 1502)
Rabbit

10mg
Beryllium sulfate
Skin sensitization and skin granulomas
 
Tinkle et al., 2003 (Document ID 1483)
Mouse
3 month
25 μL70 μg
Beryllium sulfate Beryllium oxide
Microgranulomas with some resolution over time of study
 
Intramuscular
 
 
 
 
 
 
Eskenasy, 1979 (Document ID 1496
Rabbit
35 days (injections at 7 day intervals)
10mg.ml
Beryllium sulfate
Sensitization, evidence of CBD
 
Intraperitoneal Injection
 
 
 
 
 
 
Marx and Burrell, 1973 Document ID 1450)
Guinea pig
24 weeks (biweekly injections)
2.6 mg + 10 μg dermal injections
Beryllium sulfate
Sensitization
 



Table A-3 -- Summary of Beryllium Lung Cancer Epidemiological Studies
Reference
Study type
Exposure range
Study number
Mortality ratio
Confounding factors
Study limitations
Additional comments
Beryllium Case Registry 
 
 
 
 
 
 
 
Infante et al., 1980 (Document ID 1507)
Cohort
N/D
421 cases from the BCR
SMR 2.12, 7 lung cancer deaths
Not reported
Exposure concentration data or smoking habits not reported

Steenland and Ward, 1991 (Document ID 1400)
Cohort
N/D
689 cases from the BCR
SMR 2.00 (95% CI 1.33-2.89), 28 lung cancer deaths


Included women: 93% women diagnosed with CBD; 50% men diagnosed with CBD;SMR 157 for those with CBD and SMR 232 for those with ABD.
Beryllium Manufacturing and/or Processing Plants (Extraction, Fabrication, and Processing) 
 
 
 
 
 
 
 
Ward et al., 1992 (Document ID 1378)
Retrospective Mortality Cohort
N/D
9,225 males
SMR 1.26 (95% CI 1.12-1.42), 280 lung cancer deaths

Lack of job history and air monitoring data
Employment period 1940-1969
Levy et al., 2002 (Document ID 1463)
Cohort
N/D
9,225 males
Statistically non-significant elevation in lung cancer deaths
Adjusted for smoking
Lack of job history and air monitoring data
Majority of workers studied employed for less than one year
Bayliss et al., 1971 (Document ID 1285)
Nested cohort

7,948 workers
SMR 1.06, 36 lung cancer deaths


Employed prior to 1947 for almost half lung cancer deaths
Mancuso, 1970, 1979 (Document ID 1453, 0529)
Cohort
411-43,300 μg/m[3] annual exposure (reported from Zielinsky, 1961)
1,222 workers at OH plant; 2,044 workers at PA plant
SMR 1.42 (95% CI 1.1-1.8), 80 lung cancer deaths

Partial smoking history; No job analysis by title or exposure category
Employment period from 1937-1948 with follow-up through 1967
Mancuso, 1980 (Document ID 1452)
Cohort
N/D
Same OH and PA plant analysis
SMR 1.40, 80 lung cancer deaths
No smoking adjustment
No adjustment by job title or exposure
Employment period from 1942-1948; Used workers at rayon plant for comparison
Mancuso and El Attar, 1969 (Document ID 1455)
Cohort
N/D
3,685 white males
SMR 1.49
Adjusted for age and locale
No job exposure data or smoking adjustment
Employment history from 1937-1944
Wagoner et al., 1980 (Document ID 1379)
Cohort
N/D
3,055 white males PA plant
SMR 1.25 (95% CI 0.9-1.7), 47 lung cancer deaths

Inadequately adjusted for smoking; Used national lung-cancer risk for cancer not PA
Use of 1965-1967 U.S. white male lung cancer mortality rates to generate expected numbers of lung cancers in the period 1968-1975 (which may underestimate the expected number of lung cancer deaths for the cohort) 
Sanderson et al., 2001 (Document ID 1419)
Nested case-control
 --  Average exposure 22.8μg/m[3]  --  Maximum exposure 32.4 μg/m[3] 
3,569 males PA plant
SMR 1.22 (95% CI 1.03-1.43), 142 lung cancer deaths
Smoking was found not to be a confounding factor
May not have adjusted properly for birth year or age at hire
Found association with 20-year latency
Levy et al., 2007 (Document ID 1462)
Nested case-control
Used log transformed exposure data
Reanalysis of Sanderson et al., 2001
SMR 1.04 (95% CI 0.92-1.17)
Different methodology for smoking adjustment

Found no association between beryllium exposure and increased risk of lung cancer
Schubauer-Berigan et al., 2008 (Document ID 1350)
Nested case-control
Used exposure data from Sanderson et al., 2001 
Reanalysis of Sanderson et al., 2001
Used Odds ratio: 1.91 (95% CI 1.06-3.44) unadjusted; 1.29 (95% CI 0.61-2.71) birth-year adjusted; 1.24 (95% CI 0.58-2.65) age-hire adjusted
Adjusted for smoking, birth cohort, age

 --  Controlled for birth-year and age at hire; --  Found similar results to Sanderson et al., 2001; --  Found association with 10 year latency --  "0" = used minuscule value at start to eliminate the use of 0 in a logarithmic analysis
Schubauer-Berigan et al., 2011 (Document ID 1266)
Cohort
Described in Couch et al., 2011 (updated JEM from Sanderson et al., 2001 and Chen et al., 2001)
9199 workers from 7 processing plants
SMR 1.17 (95% CI 1.08-1.28), 545 deaths
Adjusted for smoking

Male workers employed at least 2 days between 1940 and 1970
Re-evaluation of Published Studies 
 
 
 
 
 
 
 
EPA, 1998 (Document ID 0237) 
Re-analysis of Ward et al., 1992 with supporting data from Eisenbud and Lisson (1983)(Document ID 1296)
Estimated exposure range to be 100  -  1000 μg/m[3]

Elevated rate of lung cancer in overall cohort (46 cases vs 41.9 expected cases). However, not statistically significant.
Used alternative adjustment for smoking 

Characterized evidence of human carcinogenicity as "limited" but "suggestive"
Hollins et al., 2009 (Document ID 1512)
Review
Re-examination of weight-of-evidence from more than 50 publications




Found lung cancer excess risk was associated with higher levels of exposure not relevant in today's industrial settings
IARC, 2012 (Document ID 0650)
Multiple
Insufficient exposure concentration data

Sufficient evidence for carcinogenicity 
of beryllium
IARC concluded beryllium lung cancer risk was not associated with smoking

 --  Greater lung cancer risk in the BCR cohort --  Correlation between highest lung cancer rates and highest amounts of ABD or other non-malignant lung diseases --  Increased risk with longer latency --  Greater excess lung cancers among those hired prior to 1950


