As described by Nickel et al (1), there has been considerable debate and some consternation around the suggestion to rename some low-risk cancers. However, some of this may be related to a misunderstanding of the semantics of the term “cancer”. There is a general belief that “benign” and “malignant” are biologically distinct conditions, when in reality there is a continuous spectrum of increasing risk from “benign” to “good cancer” and “bad cancer”.
It is generally appreciated that the distinction between “good” and “bad” cancer is arbitrary. A 20% risk of metastasis could also be interpreted as 80% chance of not developing metastasis, so any particular risk of metastasis could be interpreted as low, intermediate or high by different patients and health professionals.
However, there is less understanding of the distinction between benign and cancer. The minimum risk threshold for a diagnosis of cancer is as subjective and arbitrary as the thresholds used to stratify cancers as low, intermediate and high risk. A low diagnostic threshold could result in overdiagnosis and overtreatment of large number of patients while using a higher threshold would risk underdiagnosis and undertreatment.
Overdiagnosis of clinically insignificant tumours as cancer could have serious physical, psychological, social and financial consequences. In addition to harming some cancer patients due to overdiagnosis and overtreatment, other patients (including those with non-cancerous diseases) may be harmed due to diversion of funds in overstretched healthcare services across the globe.
The use of arbitrary subjective diagnostic cut-offs is not unique to cancer; it is also used in several other diseases such as hypertension and diabetes. The cut-offs used to make a diagnosis of hypertension and to stratify patients into mild and severe hypertension are also arbitrary; a patient with a diastolic pressure of 100 mmHg is not at a significantly higher risk than someone with 99 mmHg even if they are in different risk categories.
In order to reduce overdiagnosis, many experts feel that the “cancer” label could be removed from some very low risk tumours, but they have been uncomfortable with labelling these low risk tumours as benign so have suggested a number of alternative terms such as “indolent lesions of low malignant potential” and “tumour of uncertain malignant potential”. However, such expressions of prognostic uncertainty could themselves induce anxiety, especially if clinical or radiological follow-up of such lesions is recommended. Use of the term “uncertain malignant potential” for low-risk tumours is particularly problematic as it may be interpreted by clinicians and patients to mean that the pathologist is uncertain whether the tumour is harmless or deadly.
Another approach could be to raise the threshold for the pathological diagnosis of cancer. As explained above, benign and cancer is a biological continuum with arbitrary cut-offs. In some instances, the criteria have been designed to identify cases that pose almost any risk to the patient, but this approach risks overdiagnosing large number of patients. If the lay population could be educated that a diagnosis of “benign” does not equate to zero risk, then some very low risk tumours could be simply re-categorised as benign thereby avoiding the anxiety induced by labels such as “cancer” or “low malignant potential”. This would amount to re-calibration of the minimum risk threshold required for the diagnosis of cancer and be analogous to changing the risk stratification thresholds for hypertension. The potential harm to a few by the delayed diagnosis of these generally indolent low risk tumours would be outweighed by the benefit to the majority, who would be spared the physical, psychological and financial effects of overdiagnosis.
REFERENCE
1. Nickel B et al. Renaming low risk conditions labelled as cancer. BMJ 2018; 362: k3322 doi: 10.1136.
Rapid Response:
Recalibrate rather than just rename?
As described by Nickel et al (1), there has been considerable debate and some consternation around the suggestion to rename some low-risk cancers. However, some of this may be related to a misunderstanding of the semantics of the term “cancer”. There is a general belief that “benign” and “malignant” are biologically distinct conditions, when in reality there is a continuous spectrum of increasing risk from “benign” to “good cancer” and “bad cancer”.
It is generally appreciated that the distinction between “good” and “bad” cancer is arbitrary. A 20% risk of metastasis could also be interpreted as 80% chance of not developing metastasis, so any particular risk of metastasis could be interpreted as low, intermediate or high by different patients and health professionals.
However, there is less understanding of the distinction between benign and cancer. The minimum risk threshold for a diagnosis of cancer is as subjective and arbitrary as the thresholds used to stratify cancers as low, intermediate and high risk. A low diagnostic threshold could result in overdiagnosis and overtreatment of large number of patients while using a higher threshold would risk underdiagnosis and undertreatment.
Overdiagnosis of clinically insignificant tumours as cancer could have serious physical, psychological, social and financial consequences. In addition to harming some cancer patients due to overdiagnosis and overtreatment, other patients (including those with non-cancerous diseases) may be harmed due to diversion of funds in overstretched healthcare services across the globe.
The use of arbitrary subjective diagnostic cut-offs is not unique to cancer; it is also used in several other diseases such as hypertension and diabetes. The cut-offs used to make a diagnosis of hypertension and to stratify patients into mild and severe hypertension are also arbitrary; a patient with a diastolic pressure of 100 mmHg is not at a significantly higher risk than someone with 99 mmHg even if they are in different risk categories.
In order to reduce overdiagnosis, many experts feel that the “cancer” label could be removed from some very low risk tumours, but they have been uncomfortable with labelling these low risk tumours as benign so have suggested a number of alternative terms such as “indolent lesions of low malignant potential” and “tumour of uncertain malignant potential”. However, such expressions of prognostic uncertainty could themselves induce anxiety, especially if clinical or radiological follow-up of such lesions is recommended. Use of the term “uncertain malignant potential” for low-risk tumours is particularly problematic as it may be interpreted by clinicians and patients to mean that the pathologist is uncertain whether the tumour is harmless or deadly.
Another approach could be to raise the threshold for the pathological diagnosis of cancer. As explained above, benign and cancer is a biological continuum with arbitrary cut-offs. In some instances, the criteria have been designed to identify cases that pose almost any risk to the patient, but this approach risks overdiagnosing large number of patients. If the lay population could be educated that a diagnosis of “benign” does not equate to zero risk, then some very low risk tumours could be simply re-categorised as benign thereby avoiding the anxiety induced by labels such as “cancer” or “low malignant potential”. This would amount to re-calibration of the minimum risk threshold required for the diagnosis of cancer and be analogous to changing the risk stratification thresholds for hypertension. The potential harm to a few by the delayed diagnosis of these generally indolent low risk tumours would be outweighed by the benefit to the majority, who would be spared the physical, psychological and financial effects of overdiagnosis.
REFERENCE
1. Nickel B et al. Renaming low risk conditions labelled as cancer. BMJ 2018; 362: k3322 doi: 10.1136.
Competing interests: No competing interests