Saving lives for the cost of an annual CT scan
This sounds alarmist, but the facts don’t lie. The National Cancer Registry, published in Oct 2016 for the period of 2007-2011, revealed that over 100,000 cancers were diagnosed in Malaysia during that time. Of these, 10,608 were new cases of lung cancer.
This means that on average, every year, 2,121 Malaysians were diagnosed with lung cancer during that time.
Overall, lung cancer is the joint number one cancer in men when it comes to incidence in Malaysia. It accounts for 15.8% of all cancers in Malaysian men, only surpassed marginally by colorectal cancer, which has an incidence of 16.3%. In women, it’s the fifth most common cancer, accounting for about 5.6% of all cases.
It is the leading cause of cancer deaths in Malaysia, as well as across the world. According to consultant cardiothoracic surgeon Dr Anand Sachithanandan, such data suggests the need for effective screening for lung cancer.
“The American College of Chest Physicians, the American Cancer Society and the American Association of Thoracic Surgery in the United States recommends screening for lung cancer,” he notes.
“What we’re dealing with is a very lethal and fairly common disease. The most common cause of cancer death in Malaysian men is lung cancer, almost one in four, which is just under 25%. In women, it’s surpassed only by breast cancer – 13% of all cancer deaths in women are due to lung cancer.
“According to the National Cancer Registry, of these 10,608 cases of cancer, they had complete data in terms of staging in 55%-60% of cases. The most alarming thing is that with lung cancer, at the time of diagnosis, only 3% were in stage 1; 7% were in stage 2; 20%-25% in stage 3; and 65%-70% in stage 4.
“Put simply, 89% of Malaysian men and 91% of Malaysian women when diagnosed, were already in stage 3 or 4.
“This is advanced disease, either locally advanced or metastatic disease. Here, automatically, the goal of treatment shifts from trying to cure the patient to palliative treatment.
“While there has been a lot of advances in chemotherapy, immunotherapy and targeted therapy, the fact remains that most of these treatments alleviate symptoms and improve quality of life, not cure.
“There are some cases where they can improve median survival, or what is called progression free survival, but unfortunately, at the end of the day, these people will still die from their lung cancer.
“And these treatments can sometimes be quite toxic and very expensive,” he says. “The only curative treatment is surgery, which is applicable in early stage disease.”
He adds: “The five-year survival for an early stage 1A lung cancer approaches 90%. Contrast that with the other end of the spectrum: stage 4B, where the five-year survival is less than 5%, and in most reported series, there’s no survival. We can see the glaring difference.”
Screening makes sense
Screening for lung cancer is not a new thing. Japan has been doing it for decades, traditionally with chest x-rays and sputum cytology. However, as time went on, they realised that such tests were not very sensitive or cost-effective.
What changed the landscape was a North American study called the NLST (National Lung Screening Trial), which was published in the New England Journal Of Medicine in 2011. This was a multi-centred prospective randomised trial involving over 30 centres in North America and Canada that looked at over 50,000 people aged between 55 and 80.
The participants were all either smokers and ex-smokers with a minimum 30-pack year history (i.e. they have smoked at least one pack of cigarettes a day for 30 years), or ex-smokers who had stopped smoking within the last 15 years.
The trial randomised these people into two groups. The participants in one group had an annual low dose CT (computed tomography) scan, while the participants in the other group had annual conventional chest x-rays. Both groups underwent their respective imaging tests for three years.
The study was able to demonstrate that death from lung cancer was reduced by 20%, while overall death from any cause was reduced by 6.7%, for those undergoing the annual CT scans.
“That became a real game-changer. It sparked an interest in screening for lung cancer around the world,” says Dr Anand. “There have been a number of trials since then; the UK Lung Screening Trial was a pilot study that has just been completed.
“Their preliminary analysis suggests that it is cost-effective, but we need to target high-risk populations. On average, all these trials and screening programmes suggest a detection rate (if you target the appropriate population) of about 2%.
“That means that out of every 100 people we screen, we pick up two lung cancer cases, of which more than 85% are early stage and amenable for curative surgery.”
Dr Anand notes that screening is a process and not an isolated test. Screening attempts to detect the disease at a pre-clinical stage when the person does not have any symptoms.
Unfortunately, in lung cancer, by the time a patient develops symptoms – persistent cough, coughing up blood, unexplained loss of weight, shortness of breath, chest pain – they are more than likely to already be in an advanced stage of the disease.
So the challenge is to try and pick up the disease early via screening.
“What the UK Lung Screening Trial did was, they used longstanding data from the Liverpool Lung Project and they identified people who had at least a 5% or more risk of developing lung cancer in the next five years – male gender, smoking history, chronic lung disease like COPD, family history of early onset lung cancer (and) a personal history of cancer.
“The other study was the Nelson study, where they also looked at a similar population – 50- to 75-year-olds and smokers – and tried to see if screening could detect the cancers earlier, and ultimately, save lives. Our challenge is whether we can simply extrapolate such data and apply it here.
“Obviously, if we target the wrong population, it will not be cost-effective, and there are potential hazards – it could lead to patient anxiety, unnecessary tests, and specifically with screening, false positives.
“This is when a scan or test suggests disease when in actual fact, there isn’t. This affects the specificity of the test. In the landmark study in North America, their false positive were quite high, around 23%.
“But the science and technology of screening has evolved quite a bit (since then). The European studies and subsequent ones used volumetric analysis – they looked at the nodule doubling time and were able to reduce significantly the false positive rate.
“In fact, their specificity was very high, something like 98%,” says Dr Anand.
He explains that when a person goes for a scan, there can be three potential outcomes.
“Obviously, normal is good. At the other end of the spectrum is someone who has something that is quite abnormal, very suspicious; at the very least, they will need close surveillance, a follow-up scan at an interval to be decided (three or four months), or if it is more sinister, a biopsy or surgery may be suggested.
“The difficulty is the intermediate group, which does not have a perfectly normal scan, but it is not typical of a cancer. Here we need to monitor closely, and we are mindful of things like patient anxiety whilst they wait for a follow-up scan.
“Those things are very important in terms of who we target,” he notes.
The scanning process
Dr Anand explains that low dose CT-scanning of the lung can often be done within five to 10 seconds. It only takes the time of a single breath and no imaging contrast is required.
“A CT scan is basically x-ray beams being directed at the body that are detected by an electronic x-ray detector rotating around the patient. It’s a form of ionising radiation and is linked to sophisticated software, which can reconstruct and create two-dimensional cross sectional views of the chest,” he says.
A valid concern is radiation from the CT scan. However, the low dose CT scans have been shown to have less than 90% of the ionising radiation of a conventional contrast CT scan.
“We are on the threshold of the next generation of ultra low dose CT scanners, where they can reduce the radiation dose to one-tenth of the existing low dose. It will end up almost similar to a chest x-ray,” says Dr Anand.
“The Cosmos study looked at the risk of long-term radiation by following a population being screened for lung cancer with annual low dose CT scans over 10 years. They found that the risk of cancer was overall, exceptionally low – 0.05%, which is five out of 10,000 of those screened after 10 years of cumulative exposure.
“That study showed that for every 108 cancers that were detected early, one person may get radiation-induced cancer.”
Nobody knows the optimum strategy when it comes to screening for lung cancer with low dose CT.
According to Dr Anand, “In the United States, they advocate annual screening until 75 or 80, or unless the person develops a co-morbidity that makes him or her a poor candidate for any definitive treatment. In the United Kingdom and Asia, we would probably be a bit more judicious.
“Because the scans are very sensitive, they can pick up something that’s a biologically insignificant tumour, something that will not progress to cause the patient a problem in their lifetime. As clinicians, we are mindful of all of this. But we have to start somewhere.”