Apple founder Steve Jobs’ recent death from pancreatic cancer focused attention on one of the fastest-spreading and deadliest malignancies for which there are no obvious red flags or screening tests.

But now, UCLA researchers may have ignited a spark of hope that a saliva test could one day detect pancreatic cancer.

People particularly dread pancreatic cancer because only 5 percent of patients are alive five years after they’re diagnosed.

Jobs was more fortunate than the majority of patients with the diagnosis, because his neuroendocrine tumor was more treatable and he survived nearly eight years from the time doctors found it.

 The new optimism centered around saliva tests begins with the basic premise that the human mouth is a virtual bacterial zoo with more than 700 species. There are good bacteria that help with digestion and immunity. There are bad bacteria linked to gum disease that also turn up in artery-clogging plaque associated with heart disease.

Writing in the journal “Gut,” Dr. James J. Ferrell and his colleagues reported finding dramatic differences in the mixtures of bacterial species in the mouths of patients with pancreatic cancer and of healthy people. Differences also emerged between the levels of particular oral bacteria in men and women with chronic pancreatitis (an inflammatory disorder and risk factor for pancreatic cancer) and healthy men and women.

The study, which appeared online today, was based on an initial comparison of bacterial species in the saliva of 10 patients whose pancreatic cancer hadn’t spread to other organs, and 10 healthy people. When the researchers analyzed quantities of various bacterial species in the various saliva samples, they found significantly more Granulicatella adiacens bacteria in the spit of cancer patients than in the spit of healthy comparison subjects. They had significantly lower levels of Streptococcus mitis and Neisseria elongata bacteria than the healthy controls.

To bolster their findings, they dug a little deeper by then examining saliva samples from 28  pancreatic cancer patients, 28 healthy people and 27 people with chronic pancreatitis. The G. adiacens levels were higher in the cancer patients than either group without cancer.

So far, they’re unable to say whether different combinations of bacteria are a cause or an effect of pancreatic cancer.  The study didn’t examine changes in oral bacteria after pancreatic cancer patients had their tumors removed, nor could it track changes in oral bacteria populations through the course of disease.

However, they said, their results suggested that saliva “is a scientifically feasible and credible biomarker source” for diseases outside the mouth and is potentially attractive because it’s non-invasive and inexpensive.

Modern medicine increasingly is finding links between infections and subsequent development of cancer. The National Cancer Institute estimates that about a fifth of cancers worldwide have their origins in bacterial, viral and other infections, with rates lower in industrialized countries and higher in developing countries where infections are less frequently treated.

Examples include the human papilloma virus (HPV), which causes the vast majority of cervical cancer, ulcer-causing H. pylori bacteria, which cause stomach cancer and infections with the hepatitis B or hepatitis C, linked to liver cancer. The Epstein-Barr virus, responsible for mononucleosis, also has been linked to some head and neck cancers most often occurring in Africa and China.