When researchers delved into how lung cancer patients were managed, they found confusion, frustration and delays.
This reflected the poor-relation status of lung cancer, a disease in which the stigma of smoking, the cause of most cases, casts a shadow that has chilled patients and their health care.
The research in Auckland and Rotorua found the median time between a patient seeking a GP's help, and diagnosis with lung cancer was two months. Those with early stage, potentially curable disease waited considerably longer, a median of 3.5 months. A quarter of patients waited more than 7.5 months.
54 per cent of lung cancer patients turned up as medical emergencies in hospital, more than twice the rate in Britain.
People diagnosed with lung cancer after arriving at an emergency department are likely to have advanced, incurable disease. One New Zealand study found that just 5 per cent of ED cases were managed with the aim of curing the cancer because it had already progressed so much.
"GPs referred patients to secondary care [hospital] by a wide variety of ways ... and were not informed of their patient's progress in secondary care," wrote research leader Wendy Stevens, of Auckland University, and her colleagues.
"Referrals sometimes got lost leading to delays as well as GP frustration."
Her group's research produced recommendations for improvement which the Northern Cancer Network's director, Dr Richard Sullivan, says were implemented to varying degrees in the Auckland region. A subsequent pilot programme to alter the "pathway of care" has improved the quality of the service and reduced delays once patients are referred to hospitals in Auckland and Northland.
"We are now implementing that pathway across the four northern DHBs."
More than 90 per cent of patients are receiving treatment within 62 days of referral.
While it will take several years to accumulate enough data on the overall effect of this improvement on patient survival rates, Dr Sullivan predicts there must be a benefit.
"Those with [potentially curable] disease, if you get them into surgery faster it would be reasonable to assume they will do better as a result."
The Stevens report found that hospital inpatients, commonly those with advanced lung cancer, received rapid investigation, whereas outpatients with early stage disease experienced longer waits.
"Those with potentially curable disease need to be diagnosed quickly as the cancer may progress while waiting."
Dr Sullivan says it is impossible to prove if a case of lung cancer has progressed while a patient is waiting for a diagnosis, "but what we do know is our curative pathway is now far quicker and provides much better quality of care".
"We are now implementing the pilot across the Auckland region. There will be pockets where it's not perfect, but we will be doing an audit in a year's time to make sure everyone is adhering to it." These developments matter greatly because lung cancer remains New Zealand's number one cause of cancer death, killing more than 1600 people a year. And our survival rate is lower than some other countries. Around 10 per cent of New Zealand lung cancer patients are alive five years after diagnosis, compared with 12 to 16 per cent in Australia and the United States.
New Zealand's lower survival rate is thought to reflect late diagnosis. Survival rates at five years can exceed 70 per cent for patients whose tumours, at diagnosis, had not spread beyond the lungs, but can be as low as 14 per cent for those diagnosed with advanced disease in which tumours have spread to other organs.
The Stevens Auckland/Rotorua report found 71 per cent had incurable disease at diagnosis.
In the New Zealand Medical Journal, she and others said this country's internationally high death rate from lung cancer - the world's leading cause of cancer deaths, killing 1.59 million people a year - is largely due to patients seeking help too late.
Part of the problem is how hard it is to distinguish a cancer cough from any other cough, and other symptoms are likewise not clear indicators of cancer. Then there is the stigma of possibly having brought the disease on yourself if you've been a long-term smoker - more than 80 per cent of lung cancer cases have smoked - leading to delays in getting to a doctor and a sense of fatalism about the likely outcome.
After barriers to referral of possible lung cancer cases to specialists were identified, the Northern Cancer Network tried to smooth the path to hospital by measures such as encouraging greater use of chest x-rays in high-risk patients and streamlining access to CT scans.
"We have done lots of education programmes around that," Dr Sullivan says. "We have gone to PHOs [primary health organisations] and GPs around high suspicion of lung cancer.
"What we haven't done is gone back to look, when a patient comes into a DHB, how many times did they present to a GP before they were referred in."
GPs were urged to find out which of their patients are at high risk for lung cancer, such as by asking and recording the smoking status of all adults.
This has been formalised as a health target for PHOs. While 21 organisations in the April-June quarter met or exceeded the target - that 90 per cent of smoking patients are offered advice and support to quit - the remaining 15, including some serving higher-smoking Maori populations, fell short.
Wide lung cancer disparities exist between Maori and non-Maori, especially for women, reflecting different rates of smoking. For females, the Maori rates of lung cancer registration and death are both more than four times greater than the non-Maori rates.
The female Maori registration rate rose by about a quarter from 2002 to 2011. Maori are also less likely than non-Maori to survive lung cancer.
In the whole adult population, the smoking rate has declined from 18.3 per cent in 2006/7 to 15.5 per cent (16.9 per cent for men, 14.1 per cent for women) in 20013/14.
The rate for Maori men was unchanged at 36.3 per cent, while for Maori women it declined from 41.8 to 37.9.
Professor Tony Blakely, an epidemiologist at Otago University in Wellington, says the lung cancer epidemic mirrors the smoking epidemic, with a two-decade time lag.
"Female lung cancer registrations are peaking, or have just peaked. They will follow males and fall, a consequence of falling smoking rates in at least the last 20 years - it takes 20 or more years before falling smoking rates are followed by falling lung cancer rates."
The men's lung cancer death rate rose steeply from 1950 to its mid-1980s peak and fell almost as quickly, but by 2010 was still about one-third higher than the women's rate.
The women's rate rose more slowly, from a lower base, and in 2010 remained almost unchanged since 1990.
The men's smoking rate was historically much higher, but dropped to meet the relatively static women's rate in the 1980s, before both fell.
Public health expert Emeritus Professor Robert Beaglehole says that because most lung cancers are caused by smoking, the efforts to reach the Government's Smokefree 2025 target of less than 5 per cent smoking prevalence are crucial.
"But the Government needs to step up. It needs to complete the standardised packaging legislation, increase taxation following the last of the 10 per cent rises next January and spend more on mass media."
Some hope the benefits of cancer screening seen in cervical and breast cancer might be applied to lung cancer, possibly with CT scanning of the lungs.
Lung cancer has been at the forefront of cancer drug research for the past decade, consistently producing new - and ever more expensive - medicines that are extending the lives of some patients, especially those with specific gene mutations targeted by the drugs.
The latest breakthrough, which is testing drug funders around the world, is a category of antibodies which, rather than attacking tumours directly, block the ability of cancer cells to hide from the body's immune system. Immune system cells then attack the tumour.
Keytruda is the first of these drugs to be registered for sale in New Zealand, but state drug-funding agency Pharmac has not yet decided whether taxpayers will pay for it. Its registration is for melanoma, but it is also being trialled in lung and other cancers.
Nivolumab is registered in the US and Europe, but not New Zealand. Atezolizumab is in trials.
"Keytruda and nivolumab have both been shown to work in a quarter of all patients with all sorts of lung cancer," says Dr Sullivan.
"There are people who are having disease control and remarkable improvements in quality of life for years on these therapies."
Four of nine patients on an Auckland trial of atezolizumab responded to the drug.
"One is out at about 19-20 months since starting and is still well. Another is at one and a half years and others at 9 to 12 months.
The international data suggests some of those people may be on it for 3 to 4 years.
"In the right patients these are really good therapies. The challenge is picking the right patients ...
"Keytruda is around $8000 per dose. You have a dose every three weeks. The data suggests you don't stop - $200,000 per year basically, so it becomes unaffordable, but maybe after six months it's done its job."
He reckons New Zealand will never be able to afford these drugs at those prices with the current practice of not setting an end date for the therapy.
"The question of duration of treatment is a really important one. And we can't afford to give them to 100 patients and spend all that money if they aren't going to be effective."