CHICAGO — Using two opposite strategies, one focused and one broad, scientists say they have made progress in taming two of the most intractable types of cancer.

The focused approach shrank tumors significantly in a majority of patients with advanced lung cancer marked by a specific genetic abnormality.

Even though the clinical trial was small (just 82 people, with no control group), the results were considered so striking for such sick patients that the study will be featured Sunday at the main session of the annual meeting of the American Society of Clinical Oncology here.

“This is a phenomenal example of finding the right patient and the right drug very early on,” said Dr. Pasi A. Janne of the Dana-Farber Cancer Institute in Boston, who was involved in the trial.

The broader strategy uses a drug that could potentially become a universal treatment for all types of cancer. It works by releasing a brake on the body’s immune system, letting the immune system attack the cancer more vigorously.

In a study of patients who had advanced melanoma, those who got an experimental drug lived a median of about 10 months, compared with 6.4 months for those in a control group. After two years, about 23 percent of those who got the drug were alive, compared with 14 percent in the control group.

Lung cancer and melanoma are among the hardest cancers to treat. So the studies are being viewed as significant advances, though far from cures.

Dr. Steven J. O’Day of the Angeles Clinic and Research Institute in Santa Monica, Calif., a lead investigator in the melanoma trial, called the result “historic,” and added, “This is the first randomized placebo-controlled trial ever to show a survival benefit in Stage 4 melanoma.”

Bristol-Myers Squibb, which sponsored the trial, is planning to apply for regulatory approval to sell the drug, ipilimumab.

The lung cancer drug, by contrast, blocks an aberrant protein called ALK that is found in only about 5 percent of non-small-cell lung tumors. But in patients whose tumors have this aberration, the drug seems to work wonders. The tumors shrank significantly in 57 percent of the 82 patients, and they remained stable in 30 percent more.

Beverly Sotir, 71, of Belmont, Mass., who has been taking the pills as part of the trial since July, said her tumors had shrunk without debilitating side effects. “For someone who’s been on chemo before, this is like a miracle drug,” she said. “You feel yourself. You look yourself.”

Pfizer, which sponsored the study, has started a more definitive trial aimed at winning approval of the drug, crizotinib.

There are caveats. The effects of crizotinib can wear off, though 72 percent of the patients in the trial were free of cancer progression for six months.

As for the melanoma drug, because it removes checks on the immune system, 10 percent to 15 percent of patients who took it in the study suffered severe side effects because their immune systems attacked their own organs. Seven patients out of 540 who got ipilimumab died from these immune effects, according to a report of the study published online Saturday by The New England Journal of Medicine.

Efforts to harness the immune system to fight cancer have suffered setback after setback. Because tumor cells are mutated forms of the body’s own cells, not an invading pathogen, they do not usually elicit a strong immune response.

But the Food and Drug Administration this year approved a “cancer vaccine” for prostate cancer called Provenge, so-called because it trains the immune system to attack the patient’s tumors. Most such vaccines focus on a single type of cancer, or are even tailored to individual patients.

Ipilimumab, by contrast, is a more general immune booster. It blocks a protein called CTLA-4 that acts as a brake on T cells, the soldiers of the immune system. It is already also being tested against lung and prostate cancer.

Still, if a tumor does not elicit a strong immune response to begin with, then just keeping the response going longer would not help much, just as lifting one’s foot from the brake usually will not make a car go faster if the accelerator is not pressed.

In at least one other melanoma trial, conventional chemotherapy drugs achieved median survival of about 10 months, the same as ipilimumab.

Dr. Charles M. Balch, a melanoma expert at Johns Hopkins who was not involved in the trial, called the results “a single, not a home run,” though he added that for this disease even a single was important.

About 68,000 Americans are expected to get melanoma this year, with 8,700 deaths, according to the American Cancer Society. The numbers have been increasing, probably because of sun exposure decades ago.

The trial involved 676 patients in the United States and 12 other countries with previously treated metastatic melanoma. They received either ipilimumab or an experimental cancer vaccine or both. Those who got ipilimumab alone did as well as those who got both, suggesting the vaccine had little effect.

Dr. Petra Rietschel of the Montefiore-Einstein Center for Cancer Care in the Bronx said melanoma experts were equally or even more excited about a drug being developed by Plexxikon and Roche that blocks a particular protein called B-RAF that is aberrant in more than half of all cases of the disease. That is similar to the approach of crizotinib, Pfizer’s lung cancer drug. They are part of a trend to genetically analyze a patient’s tumor and find drugs that block the particular genetic anomaly that drive that tumor’s growth.

Pfizer developed crizotinib to block another protein called MET. The fact that the drug also blocked ALK was considered unimportant.

But in 2007, after the clinical trial had started, Dr. Hiroyuki Mano and colleagues at Jichi Medical University in Japan reported that in a small number of lung cancers, there was a chromosome translocation that brought the gene for ALK together with the gene for another protein called EML4. That created a fusion protein that spurred tumor growth. Dr. Mano had discovered this by systematically testing all the active genes in a tumor removed from a lung cancer patient.

Pfizer turned on a dime and began enrolling lung cancer patients with this fusion protein in the trial. Japanese patients began flying to South Korea, the nearest place with trial sites.

Dr. Mano said the first Japanese patient who went was so sick — heavily dependent on oxygen tanks and unable to swallow — that he had to be taken to the airport by medical helicopter and met by an ambulance at the airport in Seoul.

Two weeks later, Dr. Mano said, he went to Seoul to check on the patient. The man no longer needed oxygen and was walking in the neighborhood each day looking for good restaurants. The patient returned to Japan and lived for several more months.

Scientists said the ALK gene aberration tends to be more frequent in younger patients and nonsmokers. Experts say that even though the drug might be useful for only 5 percent of non-small-cell lung cancer patients, that would still be about 10,000 people a year in the United States and 40,000 worldwide.

Finding drugs for each subset of tumors will take years. And cancers can mutate and become resistant to drugs blocking particular abnormalities.

Dr. James Allison, who paved the way for ipilimumab with work he did at the University of California, Berkeley, said the immune therapies might be helped by such mutations. So the targeted drugs and the immune boosting ones might work best together.

“It’s the ultimate personalized treatment for cancer,” said Dr. Allison, who is now chairman of immunology at the Memorial Sloan-Kettering Cancer Center.