Last updated: 2026-09-13

The Hong Kong Cancer Registry's Overview of Hong Kong Cancer Statistics of 2023 (August 2025) states that pancreatic cancer caused 918 deaths in 2023, 6.2% of all cancer deaths in Hong Kong, making it the fourth deadliest cancer. The same report prints a rise in pancreatic cancer deaths from 584 in 2013 to 918 in 2023, that is +57.2%; the same report attributes the rise in the number of cancer deaths for all cancers combined to population growth and ageing, which is not a separate measurement of what caused the pancreatic cancer rise. Using the age-standardised mortality rate from the Registry's CanSQS query system, the same decade runs from 3.9 to 4.4 per 100,000. The 13.2% five-year survival figure is the age-standardised relative survival of adults aged 15 to 99 diagnosed between 2018 and 2022, followed to the end of 2024, and the same calculation a decade earlier was 6.3%. In June 2022 the Government's Cancer Expert Working Group on Cancer Prevention and Screening recommended: no pancreatic cancer screening for asymptomatic persons at average risk; for persons at increased risk there is insufficient evidence for any standardised screening protocol, and those with a strong family history or specific genetic risk may seek a doctor's advice for individual assessment. These are screening recommendations; they do not apply where symptoms have already appeared and a diagnosis is needed. A rise in the number of deaths and a rise in the age-standardised mortality rate are different measures; and a survival rate cannot predict any individual's outcome.

What is pancreatic cancer? Why is it hard to find early?

Pancreatic cancer is a malignant tumour of the pancreas, a gland of the digestive system. The Department of Health's Cancer Online Resource Hub page "Pancreatic Cancer" (last revision date 28-4-2026) describes the organ and the disease this way:

The pancreas is a gland located deep in the abdomen between the stomach and the spine, connecting the duodenum. It is a digestive organ, mainly for the secretion of digestive enzymes to help digestion, and the manufacturing of insulin to monitor the blood sugar level.

The next paragraph on the same page is the official version of the "found too late" point:

Pancreatic cancer is an aggressive cancer. As the malignant tumour within the pancreas grows slowly, and hides itself deep, it is not easy to detect at its early stage. Even though for a person who has regular checks annually, it is difficult to find out if the person has got the pancreatic cancer through general tests. Therefore, pancreatic cancer is usually diagnosed at advanced stage and so delays the necessary treatments and has poor survival rate. Even the patient is able to undergo surgery to remove the tumour, his/her life span is still shorter than that of other cancer patients.

In other words, the official page attributes late detection to the location and the early symptoms, not to patients failing to go for check-ups. The Hospital Authority's Smart Patient page "Pancreatic Cancer" adds: "Even the patient is able to undergo a surgery to remove the tumour, the risk of recurrence is higher than other common cancers."

As for which type "pancreatic cancer" usually is, paragraph 5 of the Cancer Expert Working Group on Cancer Prevention and Screening (the Expert Working Group) document Recommendations on Prevention and Screening for Pancreatic Cancer of June 2022 (English only) states:

The various types of exocrine pancreatic cancer (including adenocarcinoma or ductal carcinoma, squamous cell carcinoma, adenosquamous carcinoma and colloid carcinoma) make up more than 95% of all cancers of the pancreas, while less than 5% of all pancreatic cancer cases are pancreatic neuroendocrine tumors. The vast majority (over 90%) of all diagnoses are adenocarcinoma of the pancreas.

Does "+57% in ten years" mean Hong Kong people are becoming more likely to get pancreatic cancer?

That figure is the rise in the number of deaths, not a rise in risk adjusted for the population structure, and the two have to be measured separately.

Section 2.4 of the Overview of Hong Kong Cancer Statistics of 2023 prints the 2013 to 2023 comparison, with pancreatic cancer at +57.2%; Table 2 of the same report prints the two end points: 584 deaths in 2013 (ranked 6th) and 918 in 2023 (ranked 4th). The Cancer Online Resource Hub's English page puts it this way:

In the past decade, both incidence and mortality of pancreatic cancer have significantly increased. From 2013 to 2023, the number of deaths due to pancreatic cancer jumped by 57%.

Section 2.5 of the report gives the population background for all cancers combined; it is not a separate analysis of what caused the pancreatic cancer rise. The first sentence of section 2.5:

The rise in the number of cancer deaths is attributed to population growth and ageing.

The same section goes on to say that, after accounting for demographic shifts, age-standardised cancer mortality rates have notably declined by 2.8% per year for men and 1.7% for women in the last decade, and describes this as an indication that the cancer care delivery system is making progress against cancer. ⚠️ That 2.8% / 1.7% is the age-standardised mortality rate for all cancers combined, not the figure for pancreatic cancer alone, and the two cannot be swapped.

The age-standardised figures for pancreatic cancer itself have to be looked up separately. The Hong Kong Cancer Registry's own CanSQS all-ages query system (world standard population, Segi 1960) prints year-by-year figures from 2013 to 2023: the age-standardised mortality rate for pancreatic cancer rose from 3.9 to 4.4 per 100,000, and the age-standardised incidence rate from 4.3 to 5.4.

Doing the arithmetic (this site's calculation, not a sentence written in the report): crude deaths of 584 → 918 is +57.2% (the report prints that percentage itself); over the same period the age-standardised mortality rate of 3.9 → 4.4, computed from the published figures rounded to one decimal place, is about +13%. Crude new cases of 608 → 1,049 is +72.5%; over the same period the age-standardised incidence rate of 4.3 → 5.4, by the same method, is about +26%. ⚠️ The last three percentages are this site's arithmetic on two CanSQS tables, not sentences written by the Registry, and should not be quoted as statements by that centre.

Pancreatic cancer in Hong Kong: four figures for 2013 and 2023 side by side (crude numbers and age-standardised rates). The source for each row is in the "Source" column. Age-standardised rate = ASR, using the world standard population (Segi 1960). This table selects 2013 and 2023 for comparison; the CanSQS all-ages system provides data for 1983–2023; the Overview of Hong Kong Cancer Statistics of 2023 carries a self-printed date of August 2025. Last updated: 2026-09-13.
Item20132023Ten-year changeSource
New cases (crude number)6081,049+72.5%
(this site's calculation)		Hong Kong Cancer Registry CanSQS all-ages query system, Incidence query Table 1 (www3.ha.org.hk/cancereg/allages.asp, Last updated: 2026-09-13)
Age-standardised incidence rate
ASIR (per 100,000)		4.35.4about +26%
(this site's calculation)		As above, Incidence query Table 2 (world standard population, Segi 1960)
Deaths (crude number)584918+57.2%
(printed in the report)		Overview of Hong Kong Cancer Statistics of 2023, Table 2 (document self-printed date August 2025); the same two figures also appear in the CanSQS Mortality query Table 1
Age-standardised mortality rate
ASMR (per 100,000)		3.94.4about +13%
(this site's calculation)		CanSQS all-ages query system, Mortality query Table 2 (world standard population, Segi 1960)

Within the 41-year CanSQS series from 1983 to 2023, 2023 is not the peak for any of those four indicators. The highest number of deaths was 920 in 2022, and the highest number of new cases was 1,116 in 2021; both the age-standardised mortality rate and the age-standardised incidence rate reached their highest point in that period in 2021, at 4.6 and 6.2 per 100,000 respectively.

⚠️ The report also sets its own limits on how these figures should be used. Point 1 of the "Note on the use of data" on the last page of the Overview of Hong Kong Cancer Statistics of 2023 states: "It is important to note that these figures may vary each year due to special circumstances." The same point adds that reliable insights into trends require observation over a longer period, ideally at least five years or more.

The 918 deaths and the 1,049 new cases are not follow-up results for the same group of people. The 918 comes from the Registry's Overview of Hong Kong Cancer Statistics of 2023, section 2.2 and Appendix I; the 1,049 comes from CanSQS incidence data for 2023, both sexes combined. Both are counts of people, not age-standardised.

This site's calculation: 918 ÷ 1,049 is about 87.5%, but this divides two different population statistics from the same year. It is not the chance of death for a person diagnosed, and it is not a survival rate. People who died in 2023 may have been diagnosed in earlier years.

Appendix I of the Overview of Hong Kong Cancer Statistics of 2023 also prints the 2023 pancreatic cancer deaths by sex: 495 men (5th in males, median age 71) and 423 women (4th in females, median age 73). The Appendix I footnote states: "The figures on deaths are based on deaths registered under the Births and Deaths Registration Ordinance (Cap. 174, Laws of Hong Kong)."

What exactly is the "13.2% five-year survival rate" measuring?

It is not an "overall survival rate". It is the age-standardised relative survival of a group of people with a defined age range, year of diagnosis and follow-up cut-off; and the Overview of Hong Kong Cancer Statistics of 2023 states that it rose from 6.3% a decade earlier.

The key passage in section 4.5 of the Overview of Hong Kong Cancer Statistics of 2023:

rising from 6.3% in 2010-2013 to 13.2% in 2018-2022

The full sentence says that pancreatic cancer, despite having the lowest survival rate, showed a significant improvement in five-year survival. The highlights page of the same report also lists pancreatic cancer under "Most Improved Prognosis", printing "Pancreatic cancer (109.5% relative change)", that is a relative change of +109.5%.

That 13.2% carries a whole set of conditions with it, and dropping any one of them turns it into a different statement:

  • How it is measured: age-standardised relative survival, not observed survival. Section 4.1 of the report states that this measure is commonly used to assess cancer survival while controlling for demographic changes.
  • Age: adults aged 15 to 99.
  • Year of diagnosis: 2018 to 2022.
  • Followed to when: the end of 2024.
  • Size of the study: the whole study covers over 370,000 adults diagnosed with 20 major solid cancers between 2010 and 2022; the 13.2% is only the pancreatic cancer result for the 2018–2022 group within it.
  • The scope statement printed on the chart itself: the note to Figure 6 of the report reads "5-year relative survival rates (%) for cases diagnosed during 2018-2022 are shown."

How does it compare with other cancers? The highlights page of the same report prints: "Poor Prognosis: Pancreatic cancer (13.2%), Oesophageal cancer (16.3%), and Gallbladder & extrahepatic bile duct cancer (17.7%)", and "Best Prognosis: Thyroid cancer (92.4%), Female breast cancer (86.0%), and Prostate cancer (84.2%)". For all cancers combined, section 4.2 states that "the 5-year relative survival improved from 49.1% in 2010-2013 to 55.7% in 2018-2022".

The most important sentence is the one the report writes itself. Point 2 of the "Note on the use of data" on the last page of the Overview of Hong Kong Cancer Statistics of 2023:

Survival statistics are derived from data on patients diagnosed in the past and may not reflect each individual's situation.

⚠️ That is a limit written by the body that published the figure. This article treats it accordingly: it makes no prediction, and can make none, about the outcome for any individual reader or their family.

The 13.2% is not broken down by stage of pancreatic cancer. Section 4.6 of the Registry's 2023 Overview notes that age and stage at diagnosis influence cancer survival. The pancreatic cancer summary in that report, the Registry's "Cancer Statistics" page and the six stage-specific survival reports listed on its "News" page do not set out stage-specific survival for pancreatic cancer; that limit applies only to those three sources and is not a statement that no official publication contains it.

What is the difference in screening advice for people at average risk and at increased risk?

The Hong Kong recommendation distinguishes asymptomatic persons at average risk from persons at significantly increased risk; diagnostic assessment for someone who already has symptoms is a separate matter.

The "Screening" section of the Cancer Online Resource Hub page "Pancreatic Cancer" (last revision date 28-4-2026), in the official English:

Based on the available international and local scientific evidence, the Government's Cancer Expert Working Group on Cancer Prevention and Screening has made the following recommendations on pancreatic cancer screening for the local population:
For persons at average risk
Screening for pancreatic cancer (including screening by serum biomarker CA19-9) is not recommended in asymptomatic persons at average risk.
For persons at increased risk
There is currently insufficient evidence to recommend screening of pancreatic cancer for persons at increased risk by any standardised protocol. Persons with strong family history of pancreatic cancer, specific genetic syndromes, or carrying genetic susceptibility traits that put them at significantly increased risk of pancreatic cancer may consider seeking advice from doctors for individual assessment.

When and by whom was this recommendation adopted? The Expert Working Group document Recommendations on Prevention and Screening for Pancreatic Cancer (For Health Professionals) carries a self-printed date of June 2022, and paragraph 23 states that the recommendations "were endorsed by the Cancer Coordinating Committee at its 17th meeting on 7 June 2022".

The reasons behind the recommendation are in that June 2022 professional document:

  • Paragraph 11 gives as one of the reasons "the absence of a suitable non-invasive test", in a sentence that also cites the relatively low incidence of pancreatic cancer in the general population and a lack of data on various aspects as demonstrated by the USPSTF's systematic review.
  • Paragraph 16: "More importantly, there was a lack of evidence on whether screening HRIs would save lives, while available evidence supporting screening and surveillance was limited to observational studies." (HRIs is the document's abbreviation for high-risk individuals.)
  • Paragraph 21 also states that local research is lacking: "Additional research is needed to better understand the risk factors or causes of pancreatic cancer in local Chinese population." The same paragraph notes that there exists a possibility that screening in the general population might lead to over-diagnosis and over-treatment, and stresses that awareness of symptoms and signs, timely investigations and referral in the primary care setting play a key role in early detection to improve the prognosis.
  • Paragraph 12, on the international picture, states that in the United Kingdom the NHS "does not have a national programme to screen the general population for pancreatic cancer". The same paragraph also records that the IARC of the World Health Organization, the USPSTF and Cancer Australia do not recommend screening the general population.

⚠️ About the blood marker CA19-9. The "Cancer diagnosis" section of the Cancer Online Resource Hub states: "Blood marker carbohydrate antigen 19-9 (CA 19-9) is elevated in some patients with pancreatic cancer. It may play a role in disease monitoring and prognosis, together with radiological/imaging and clinical data. However, it can be elevated in other biliary conditions such as cholangitis, biliary obstruction of other aetiologies, and some other types of cancers." The same marker is expressly excluded by the official page from screening use in persons at average risk, and at the same time described as an aid in monitoring disease — these two things are not in conflict, because the uses are different.

What symptoms and risk factors do the official pages list?

Someone who has symptoms is not in the "asymptomatic screening" question. The Centre for Health Protection's pancreatic cancer public leaflet says to consult a doctor as soon as possible if any of the listed symptoms develop. A symptom list is not a diagnostic standard, and it does not set a number of days to wait before assessment.

(1) The Cancer Online Resource Hub page "Pancreatic Cancer" (last revision date 28-4-2026), six items:

Should a patient have the following symptoms, he/she may get the disease:
Sustained pain in the upper abdomen, which is not related to eating or drinking, and the pain may extend to the back.
Loss of appetite, nausea, vomiting, indigestion, physogastry, and other intestine and stomach problems.
Jaundice, itching skin and clay coloured stools.
Drastic loss of weight in a short time.
Fixed, hard lumps in upper abdomen.
Ascites.

(2) The Centre for Health Protection leaflet Pancreatic Cancer Prevention and Screening also lists tiredness, dark-coloured urine, pale and greasy stool, and "Blood clots in large vein, causing swelling and pain of leg due to deep venous thrombosis". The pale stool overlaps with the clay coloured stools in the list above; greasy is an additional characteristic. The leaflet is published in both Chinese and English.

After the symptom list the English leaflet states:

You should consult a doctor as soon as possible if you develop any symptoms.

That advice to seek medical attention does not stop applying because screening is not recommended for persons at average risk.

(3) Risk factors. The "Prevention" section of the Cancer Online Resource Hub page "Pancreatic Cancer" lists: "Smoking", "Overweight or obesity", "Consumption of red or processed meat", "Heavy alcohol consumption", "Elderly males", "Family history of pancreatic cancer and carrier of certain inherited genes", and "History of certain diseases (e.g. diabetes mellitus, chronic or hereditary pancreatitis)", introduced by the words "While it is not clear what causes pancreatic cancer, some risk factors identified are:". A separate section on the same page, "Who have higher chance of having Pancreatic Cancer?", prints a not-identical list of 10 items, among them "Smokers' chance of getting the disease is 2 to 3 times higher than that of non-smokers." The seven risk factors of the "Prevention" section are set out above; from "Who have higher chance of having Pancreatic Cancer?" only the smoking item is extracted here.

Paragraph 6(a) of the Expert Working Group document says of smoking: "11-32% of all pancreatic cancer might be attributable to tobacco smoking". ⚠️ What that paragraph cites is a summary review of meta-analytical studies, not a figure measured locally in Hong Kong. Paragraph 6(a) of that document gives no local attributable fraction for Hong Kong, so the 11–32% cannot be treated as a Hong Kong figure.

If something is suspected, how does the public system work, and what does it cost?

This section is about administrative arrangements, not clinical judgement. The Hospital Authority's "Specialist Out-patient" page states:

Hong Kong citizens are encouraged to visit Family Medicine Clinic (FMC) or private family doctor when they need medical consultations. FMCs or their family doctors will also consider making referrals to specialist outpatient clinics of Hospital Authority for specialist consultation, treatment and investigation, if necessary.

To book a new-case appointment you must bring your "HK Identity Card (or a valid identity document)", a "Referral letter from a local registered medical practitioner issued in the last three months", and "Residential address information". After referral, the Hospital Authority's "Triage Arrangements of Specialist Out-patient Clinics" page states: "Referrals are usually first screened by a nurse and then by a specialist of the relevant specialty for classification into the following categories", namely Urgent Case (Priority 1), Semi-urgent Case (Priority 2) and Stable Case (Routine), and "Patients classified as urgent and semi-urgent would be arranged to have the consultation within 2 weeks and 8 weeks respectively as far as possible." The triage page also keeps an arrangement for a condition that worsens: "If a patient's condition deteriorates before the appointment, he or she may contact the clinic concerned and request for an earlier appointment. If the condition is acute, the patient could also seek immediate treatment at the Accident and Emergency (A&E) departments. Arrangement for the patients would be made as necessary."

Published specialist out-patient waiting times are not pancreatic cancer diagnosis times. The Hospital Authority's specialist out-patient waiting time dataset for 1 July 2025 to 30 June 2026 covers eight specialties — paediatrics, medicine, surgery, gynaecology, ophthalmology, psychiatry, ear, nose and throat, and orthopaedics and traumatology; it does not set out any pancreatic-cancer-specific pathway or clinical oncology. The median for stable new cases in surgery cannot stand for the waiting time of someone with suspected cancer or for an urgent case, as cases are triaged by clinical condition.

What follows are the Hospital Authority's charges for eligible persons, not a pancreatic cancer investigation or treatment package. The "Fees and Charges" page states that the fees have been effective since 1 January 2026 and are charged as per the Gazette: specialist clinic $250 per attendance, and $20 per drug item, each chargeable unit covering four weeks. Family medicine clinic $150 per attendance, and $5 per drug item, likewise per four weeks. Day procedure and treatment at a clinical oncology clinic or renal clinic / centre is $250 per attendance. Self-financed drugs and non-standard charge items are arranged separately; the above is not a total annual cost.

Accident and Emergency is $400 per attendance, and patients triaged as Category I (critical) and Category II (emergency) are fee exempted; an acute general bed is $300 per day, the public ward maintenance fee covering consultation, examinations, general prescriptions, vaccines, general nursing and meal, and if a patient is discharged within the same date of admission one day's maintenance fee is levied. Children under twelve years of age and babies who cannot be discharged at the same time as their mothers are charged half the maintenance fees appropriate to the type of bed occupied, with all other fees the same as those for adults (notes N3, N4 and N7 on the same page).

The definition of eligible persons is in item (B) of that page, and covers holders of a Hong Kong Identity Card meeting the stated conditions, children who are Hong Kong residents and under 11 years of age, and other persons approved by the Chief Executive of the Hospital Authority. Other people are charged under a different set of rates. The Hospital Authority also operates medical fee waivers and an annual cap on designated public healthcare charges; what each actually covers is governed by the rules of the scheme concerned.

Common questions

  • Where does pancreatic cancer rank in Hong Kong? Section 2.2 of the Overview of Hong Kong Cancer Statistics of 2023 and the Cancer Online Resource Hub's English page both show that in 2023 pancreatic cancer was the fourth leading cause of cancer deaths, with 918 deaths, 6.2% of all cancer deaths. That is a mortality ranking, not an incidence ranking.
  • Deaths rose 57% in ten years — does that mean risk rose by 57%? It cannot be read that way. Section 2.5 of the report attributes the rise in the number of cancer deaths for all cancers combined to population growth and ageing, and is not a separate analysis of what caused the pancreatic cancer rise; over the same period the age-standardised mortality rate for pancreatic cancer went from 3.9 to 4.4 per 100,000 (about +13% computed from the published figures rounded to one decimal place). Both series are printed in this article — see the table above.
  • Five-year survival is 13.2% — does that mean I or my family member have a 13.2% chance? The Registry that published the figure states on the last page of the Overview of Hong Kong Cancer Statistics of 2023: "Survival statistics are derived from data on patients diagnosed in the past and may not reflect each individual's situation." The 13.2% is the age-standardised relative survival of adults aged 15 to 99 diagnosed in 2018–2022 and followed to the end of 2024; the same calculation a decade earlier gave 6.3%.
  • Can a CA19-9 blood test find pancreatic cancer early? The "Screening" section of the Cancer Online Resource Hub states: "Screening for pancreatic cancer (including screening by serum biomarker CA19-9) is not recommended in asymptomatic persons at average risk." The same page also states that, together with radiological/imaging and clinical data, the marker may play a role in disease monitoring and prognosis, but that it can be elevated in other biliary conditions and in some other types of cancers.
  • I have a family history — what can I do? The "Screening" section of the Cancer Online Resource Hub page "Pancreatic Cancer" puts it this way: "There is currently insufficient evidence to recommend screening of pancreatic cancer for persons at increased risk by any standardised protocol", while persons with a strong family history, specific genetic syndromes or genetic susceptibility traits that put them at significantly increased risk "may consider seeking advice from doctors for individual assessment". That is the individual-assessment arrangement that page sets out for persons at significantly increased risk.
  • Upper abdominal pain, indigestion, weight loss — how long should I wait? The Centre for Health Protection's pancreatic cancer leaflet advises consulting a doctor as soon as possible if symptoms develop, and sets no number of days. The Hospital Authority's specialist out-patient guidance sets out referral by a family doctor as necessary; if the condition deteriorates while waiting, the patient may contact the clinic to request an earlier appointment, and in an acute condition may attend Accident and Emergency. These are official arrangements for different situations, not a single pathway.

Related articles: 〈Colorectal Cancer Screening Programme〉; 〈Is a full-body MRI check-up worth it〉; 〈Common indicators on a blood test report〉

What this article does not state

  • This article does not use population survival to predict an individual outcome, and does not divide the number of deaths by the number of new cases to produce a chance of death for a person diagnosed.
  • This article does not give survival by stage of pancreatic cancer. The sources limited above are the 2023 Overview, the Registry's "Cancer Statistics" page and its "News" page; that limit does not mean no official document has published such figures.
  • This article does not give a pancreatic-cancer-specific waiting time or any private investigation or treatment quotation; the Hospital Authority dataset for the eight specialties above sets out no pancreatic-cancer-specific pathway.
  • The screening recommendations are limited to the risk groups listed and to the asymptomatic situation; this article cannot judge any individual condition or its urgency. The Traditional Chinese edition's Chinese renderings of English-language sources are that edition's own summaries, not official Chinese translations.
  • This is the English edition. The Traditional Chinese edition is the authoritative version of this article. Quotations above are reproduced from the official English text published by the Government of the Hong Kong Special Administrative Region and by the bodies named, not translated by this site.

Sources

English-language sources are quoted in their original English; the Chinese explanations in the Traditional Chinese edition are that edition's own summaries rather than official Chinese translations. Official figures and arrangements may be updated, and what the bodies concerned publish governs. This article is health information, not medical advice, and it cannot judge whether any individual condition is urgent. The Hospital Authority's guidance states that a patient whose condition becomes acute while waiting for a specialist out-patient service should seek immediate treatment at an Accident and Emergency department.