Slide 1

Cancer clusters kNOw Cancer - 2015 Constructively dealing with concerns about cancer clusters Tim Driscoll University of Sydney

Slide 2

Cancer clusters kNOw Cancer - 2015 Aims Provide an overview of cancer clusters Consider the main issues Recommend an approach Propose some principles Case study

Slide 3

Cancer clusters kNOw Cancer - 2015 What is a cancer cluster? Various definitions An unusually high number or rate of cancer - (given the relevant population and time period) Usually describes one type of cancer, but may include different types of cancer Should NOT imply a connection with a single cause, but often this is what is meant by people using the term

Slide 4

Cancer clusters kNOw Cancer - 2015

Slide 5

A few concepts about cancer Common disease Usually more common with age Long latency Usually has one or more known risk factors Has a random component differences in rates often due to chance

Slide 6

Cancer clusters kNOw Cancer - 2015 Why is it so hard?? Most cancers can be caused by more than one type of exposure The same exposure can occur in different settings Long latency obscures the connection to exposure Usually no way to determine the cause in an individual case Random doesnt mean even

Slide 7

Cancer clusters kNOw Cancer - 2015 What is chance?

Slide 8

Cancer clusters kNOw Cancer - 2015 What is chance? Random co-location of different risk factors over time and place Random occurrence of cases over time and place

Slide 9

Cancer clusters kNOw Cancer - 2015

Slide 10

Slide 11

Slide 12

Slide 13

When do concerns arise?

Slide 14

Cancer clusters kNOw Cancer - 2015 When do concerns arise? Cases of the same cancer type Lots of cases Young people Concerning exposures Not giving initial concerns appropriate consideration Industrial relations or community issues

Slide 15

Cancer clusters kNOw Cancer - 2015 Should clusters be investigated? To summarise, I would recommend that we spend less time reacting to reports of disease clustering, less time trying to detect general patterns of disease clustering, and less time developing new methods to conduct these activities. Ken Rothman, 1990 It is fair to state that extensive efforts to find causes of community cancer clusters have not been successful. There are fundamental shortcomings to our current methods of investigating community cancer clusters. Goodman et al, 2012 Critical Reviews in Toxicology2012;42(6):474490 AJE 1990;132(Suppl 1):S6-S13

Slide 16

Cancer clusters kNOw Cancer - 2015 Why bother?

Slide 17

Cancer clusters kNOw Cancer - 2015

Slide 18

Cancer types involved Breast Bladder Colon Kidney Myeloma Multiple 511117511117

Slide 19

Cancer clusters kNOw Cancer - 2015 Settings involved Office setting University Industry Public service Research institute Residential care Art Gallery Fire station School 322222111322222111

Slide 20

Cancer clusters kNOw Cancer - 2015 When do concerns arise? Cases of the same cancer type Lots of cases Young people Concerning exposures Not giving initial concerns appropriate consideration Industrial relations or community issues

Slide 21

Cancer clusters kNOw Cancer - 2015 When should concerns be raised?

Slide 22

Cancer clusters kNOw Cancer - 2015 When should concerns be raised? Number of cases? Type of cancer? Ages of affected persons? Type of exposures? Never?

Slide 23

Cancer clusters kNOw Cancer - 2015 When should concerns be raised? Number of cases?...........Usually need more than just a few Type of cancer? Ages of affected persons? Type of exposures?

Slide 24

Cancer clusters kNOw Cancer - 2015 When should concerns be raised? Number of cases?...........Usually need more than just a few Type of cancer?...............Rare or unusual cancers; same type Ages of affected persons? Type of exposures?

Slide 25

Cancer clusters kNOw Cancer - 2015 When should concerns be raised? Number of cases?...........Usually need more than just a few Type of cancer?...............Rare or unusual cancers; same type Ages of affected persons?.....Young people Type of exposures?

Slide 26

Cancer clusters kNOw Cancer - 2015 When should concerns be raised? Number of cases?...........Usually need more than just a few Type of cancer?...............Rare or unusual cancers; same type Ages of affected persons?.....Young people Type of exposures?..........Known carcinogens; known connection to identified cancers

Slide 27

Cancer clusters kNOw Cancer - 2015 Challenges

Slide 28

Cancer clusters kNOw Cancer - 2015 Challenges Lack of information about exposure Uncertainty about case definition Incomplete case identification Uncertainty about population at risk

Slide 29

Cancer clusters kNOw Cancer - 2015 Challenges Interpretation of statistical tests Public perception that there must be a problem Context Other agendas Ethics restrictions / requirements

Slide 30

Cancer clusters kNOw Cancer - 2015 Response

Slide 31

Cancer clusters kNOw Cancer - 2015 Response Prompt response Explain the challenges Emphasis on exposures and case characteristics rather than on rates Involve all interested parties (reference group) Regular feedback

Slide 32

Cancer clusters kNOw Cancer - 2015 The extent of investigation can vary The extent of investigation depends on the specific circumstances Sometimes a discussion and a brief site visit is all that is needed Sometimes a detailed investigation is clearly indicated Always start by talking at an early stage to the concerned people

Slide 33

Cancer clusters kNOw Cancer - 2015 Two stage approach Stage 1: review of past and current exposures - ALL carcinogens, with a focus on the relevant ones Stage 2: epidemiological analysis of cancer cases - emphasis on case characteristics Stage 1 and Stage 2 usually overlap Good communication throughout

Slide 34

Cancer clusters kNOw Cancer - 2015 Driscolls four principles for investigating cancer clusters

Slide 35

Cancer clusters kNOw Cancer - 2015 Important principle 1 Cancer cluster investigations are socio-scientific phenomena - Good science is not enough - Good science AND good communication and consultation is required

Slide 36

Cancer clusters kNOw Cancer - 2015 Important principle 2 If no concerning exposures are found but there is a high rate....................... .. it is almost certain that the high rate was due to chance If still concerned, need to study another workplace/community with a similar exposure

Slide 37

Cancer clusters kNOw Cancer - 2015 Important principle 3 If rate is not high but concerning exposures are found................................... .Fix the exposures!

Slide 38

Cancer clusters kNOw Cancer - 2015 Important principle 4 When concerns about a cancer cluster arise, the cancers nearly always turn out not to be (or almost certainly not to be) related to work (or community) exposures BUT....it is important to still investigate properly

Slide 39

Cancer clusters kNOw Cancer - 2015 The key aspects

Slide 40

Cancer clusters kNOw Cancer - 2015 The key aspects Focus on the exposures and the concerns of the individuals Listen to ALL concerns and address them to the extent possible Communicate early and often Instead, we should focus more on exposure assessment and, where indicated, cleanup. Ken Rothman 1990 A duty of care was seen to be met when the investigation was extended beyond carcinogens that would account for the cluster to all carcinogens that were worrying those affected. Bernard Stewart 2007 AJE 1990;132(Suppl 1):S6-S13 MJA 2007; 187: 178180

Slide 41

Cancer clusters kNOw Cancer - 2015 Advantages and disadvantages

Slide 42

Cancer clusters kNOw Cancer - 2015 Advantages of investigation Provide reassurance that carcinogenic exposures in the setting are CURRENTLY not present or are well controlled OR Identify exposures that are not well controlled so they can be FIXED

Slide 43

Cancer clusters kNOw Cancer - 2015 Advantages of investigation 2 Provide insight into whether PREVIOUS carcinogenic exposures in the setting were likely or not likely to be responsible for identified cancers Provide guidance regarding whether the identified occurrence of cancer is or isnt unusual Provide a forum for concerns to be addressed There is little choice - the concerns rarely disappear!

Slide 44

Cancer clusters kNOw Cancer - 2015 Disadvantages of investigation In many cases the rate of cancer will not be shown to be higher (i.e. there is no cluster) A causal connection to an exposure is very rarely identified Commonly costly in terms of time and resources The final outcome is commonly not definitive

Slide 45

Cancer clusters kNOw Cancer - 2015 Investigation of a reported cluster of cancer cases at the National Gallery of Australia Final report Tim Driscoll Gary Foster Felicity Driscoll September 2008

Slide 46

Cancer clusters kNOw Cancer - 2015 National Gallery - background Concerned staff Management perceived as slow to react Initial investigation very limited Staff more concerned and unhappy!

Slide 47

Cancer clusters kNOw Cancer - 2015 National Gallery what did the study team do? Set up a reference group Reviewed relevant scientific literature Workplace investigation for carcinogens Interviewed workers (but not all cases) Number, rate, type and characteristics of cancer in current and past workers

Slide 48

Cancer clusters kNOw Cancer - 2015 National Gallery - exposures IARC Group 1 Asbestos Benzene Cadmium Environmental tobacco smoke Ethylene oxide Formaldehyde Radium Wood dust X-rays

Slide 49

Cancer clusters kNOw Cancer - 2015 National Gallery - exposures IARC Group 1 Asbestos Benzene Cadmium Environmental tobacco smoke Ethylene oxide Formaldehyde Radium Wood dust X-rays IARC Group 2A Diesel fumes Epichlorohydrin Polycyclic aromatic hydrocarbons Tetrachloroethylene

Slide 50

Cancer clusters kNOw Cancer - 2015 National Gallery - exposures IARC Group 1 Asbestos Benzene Cadmium Environmental tobacco smoke Ethylene oxide Formaldehyde Radium Wood dust X-rays IARC Group 2A Diesel fumes Epichlorohydrin Polycyclic aromatic hydrocarbons Tetrachloroethylene IARC Group 2B Carbon black Carbon tetrachloride Cobalt Dichloromethane Dichlorvos Extremely low-frequency electro- magnetic fields Magenta Potassium bromate Synthetic mineral fibres Welding fumes

Slide 51

Cancer clusters kNOw Cancer - 2015 National Gallery - exposures IARC Group 1 Asbestos Benzene Cadmium Environmental tobacco smoke Ethylene oxide Formaldehyde Radium Wood dust X-rays IARC Group 2A Diesel fumes Epichlorohydrin Polycyclic aromatic hydrocarbons Tetrachloroethylene IARC Group 2B Carbon black Carbon tetrachloride Cobalt Dichloromethane Dichlorvos Extremely low-frequency electro- magnetic fields Magenta Potassium bromate Synthetic mineral fibres Welding fumes

Slide 52

Cancer clusters kNOw Cancer - 2015 National Gallery what was found? 57 current and former workers diagnosed with cancer Types with the highest numbers were the most common community cancer types Little to suggest common exposures of cases (except security guards) All cancers..........................................No increase Lung cancer........................................40% increase (very likely due to chance) Bowel cancer.......................................6% increase (very likely due to chance) Bowel cancer (security guards)..........Increased risk (very likely due to individual risk factors)

Slide 53

Cancer clusters kNOw Cancer - 2015 National Gallery what was found? Lots of carcinogens No exposures at a level that would meaningfully increase risk

Slide 54

Cancer clusters kNOw Cancer - 2015 National Gallery what was concluded? Very unlikely that any of the cancers identified in Gallery staff members were related to exposures experienced while working in the Gallery building No further investigation of the issue considered necessary Re-design the loading dock

Slide 55

Cancer clusters kNOw Cancer - 2015 National Gallery what was the outcome? The loading dock was demolished as part of renovations The new loading dock is very well designed No further issues re cancer The gallery's deputy director, Alan Froud, said he was pleased the report put to rest the long-running concerns. The former head of conservation and a long-term critic of the gallery's health and safety practices, Bruce Ford, said he was pleased it had been cleared. But the cancer study had not addressed his broader concerns about the health of the building. SMH 3/10/2008

Slide 56

Cancer clusters kNOw Cancer - 2015 Jilalan rail yard: assessment of exposures and concerns regarding cancer amongst current and former workers Final report Tim Driscoll ELMATOM Pty Ltd April 2010

Slide 57

Cancer clusters kNOw Cancer - 2015

Slide 58

Guidance in Australia NHMRC: https://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/ps0006_statement_cancer_clusters.pdf https://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/ps0006_statement_cancer_clusters.pdf Cancer Council: http://www.cancer.org.au/news/blog/risks/what-are-cancer-clusters.html http://www.cancer.org.au/news/blog/risks/what-are-cancer-clusters.html Queensland Health: http://www.health.qld.gov.au/epidemiology/documents/faq-cancer-cluster.pdf http://www.health.qld.gov.au/epidemiology/documents/faq-cancer-cluster.pdf Vic Health: http://www.betterhealth.vic.gov.au/bhcv2/bhcarticles.nsf/pages/Disease_clusters

Slide 59

Cancer clusters kNOw Cancer - 2015 Change what you can change

Slide 60

Cancer clusters kNOw Cancer - 2015 Change what you can change Stop smoking Exercise regularly Eat plenty of vegetables and fruit Maintain appropriate weight Avoid excessive alcohol consumption Restrict sun exposure

Slide 61

Cancer clusters kNOw Cancer - 2015 Conclusions

Slide 62

Cancer clusters kNOw Cancer - 2015 Conclusions 1 Cancer is a common disease Cancer clusters: are expected due to random variation are usually not caused by occupational or environmental exposures are very rarely due to unknown exposures should not be dismissed without investigation of some sort good communication is essential

Slide 63

Cancer clusters kNOw Cancer - 2015 Conclusions 2 Prompt response Explain the challenges Emphasis on exposures and case characteristics rather than on rates Involve all interested parties Regular feedback Change what you can change

Slide 64

Cancer clusters kNOw Cancer - 2015