Immunotherapy releases the brakes that a tumor places on the immune system, and checkpoint inhibitors have produced five-year survivors in melanoma, lung, kidney, bladder and other cancers that once had none. It covers how the drugs work, why they help some patients and not others, which cancers respond, what the five-year trials show, the course of infusions, the immune-related side effects and their warning signs, and how patients from abroad arrange it in Istanbul.
About This Department
Medical oncology
The immune system can kill cancer. For decades the cancer stopped it. Drugs that release the brake have turned some of the deadliest cancers into diseases that half of patients survive for years. For some. Knowing which is the whole art.
Immunotherapy treats cancer by turning the patient's own immune cells against it rather than by poisoning the cancer directly. The drugs that made it work, called checkpoint inhibitors, block the signals a tumor uses to switch off the T cells that would otherwise destroy it. In the trial that established the approach, patients with advanced melanoma treated with two checkpoint inhibitors together had a five-year survival of 52 percent, against 26 percent with the older single drug, and no lasting loss of quality of life (Larkin and colleagues, New England Journal of Medicine, 2019). In lung cancer with high levels of the PD-L1 marker, a single checkpoint inhibitor doubled five-year survival over chemotherapy, from 16.3 to 31.9 percent (Reck and colleagues, Journal of Clinical Oncology, 2021). Serious side effects come from the immune system attacking healthy organs, and deaths from them occur in 0.36 to 1.23 percent of patients depending on the regimen (Wang and colleagues, JAMA Oncology, 2018). This page explains what immunotherapy is, why it works in some patients and not others, which cancers it treats, what the trials show, what a course involves, the side effects, how it is combined with other treatments, and how patients from abroad arrange it at Biruni Hospital in Istanbul.
52
Percent of patients with advanced melanoma alive at five years on nivolumab plus ipilimumab, against 26 percent on ipilimumab alone
31.9
Percent five-year survival with pembrolizumab in lung cancer with PD-L1 of 50 percent or more, against 16.3 percent with chemotherapy
0.36
Percent of patients on a single anti-PD-1 drug who died of a side effect, across 112 trials and 19,217 patients
Free
Review of your pathology, biomarker results and scans by the medical oncology team
T cells patrol the body looking for cells that have gone wrong, and they kill what they find. Cancer cells go wrong by definition, and many of them display fragments on their surface that a T cell should recognize. They survive anyway, because they have learned to press a switch on the T cell, a checkpoint, that tells it to stand down. The switch exists to stop the immune system attacking healthy tissue. The tumor borrows it.
Three kinds
1
Checkpoint inhibitors
Antibodies given by infusion every two to six weeks that block the switch, most often PD-1 on the T cell, PD-L1 on the tumor, or CTLA-4. Pembrolizumab, nivolumab, atezolizumab, durvalumab, ipilimumab and their newer relatives. This is what nearly everyone means by immunotherapy, and what this page describes.
2
Cell therapies
The patient's own T cells removed, engineered or expanded outside the body, and returned. CAR-T cell therapy for blood cancers is the established form. The hospital describes it on its own page.
3
Older and narrower forms
BCG instilled into the bladder for early bladder cancer, interferon and interleukin for a few cancers, and therapeutic vaccines, which remain experimental for most tumors.
Overall survival at 5 years was 52 percent in the nivolumab-plus-ipilimumab group and 44 percent in the nivolumab group, as compared with 26 percent in the ipilimumab group. No sustained deterioration of health-related quality of life was observed (Larkin and colleagues, 2019).
What that sentence means
Before 2011, a patient with melanoma that had spread lived a median of six to nine months, and almost nobody lived five years. The trial quoted above enrolled 945 such patients. More than half of those who received the two-drug combination were alive five years later, most of them off treatment, and the survival curve had gone flat, which in cancer medicine is the shape of a cure. The same shape has since appeared in lung, kidney, bladder, head and neck and several other cancers, in a minority of patients each time. Immunotherapy changed the question from how long to whether.
Why it works for some and not others
A checkpoint inhibitor does nothing to the cancer. It works only if T cells that recognize the cancer already exist and are being held back. Where they do not exist, releasing the brake releases nothing.
The markers
Three tests predict, imperfectly, which tumors will respond. PD-L1 staining on the tumor cells, expressed as a percentage, which is high in cancers that are actively using the switch and which decided the lung cancer trial below. Microsatellite instability, or a deficient mismatch repair system, found in a small proportion of bowel, uterine, stomach and other cancers, which makes the tumor produce so many abnormal proteins that the immune system cannot miss it. And tumor mutational burden, a count of mutations per stretch of DNA, high in cancers caused by sunlight and tobacco. A tumor with none of these can still respond, and one with all three can still fail, which is the reason the trials matter more than the markers and the reason the oncologist reads both.
Which cancers it treats
Melanoma first. Then the list grew.
Which cancers respond to immunotherapy?
Melanoma, non-small cell lung cancer, kidney cancer, bladder and urinary tract cancer, head and neck cancer, Hodgkin lymphoma, Merkel cell and cutaneous squamous cell skin cancers, liver cancer, esophageal and stomach cancer, triple-negative breast cancer, cervical cancer, endometrial cancer, mesothelioma and small cell lung cancer, in each case for a defined stage and often only with a marker present. Any cancer with microsatellite instability, wherever it started. The list changes every year as trials report. Cancers that respond poorly so far include pancreatic cancer, prostate cancer, most bowel cancers without microsatellite instability, and brain tumors, and a patient with one of these who is offered immunotherapy outside a clinical trial should ask on what evidence.
For most of the cancers above, immunotherapy is given first, alone or with chemotherapy, for disease that has spread. For several it is also given before or after surgery for earlier stages, to reduce the chance of the cancer returning, and this is where the field is moving fastest. A patient with an early cancer should ask whether a checkpoint inhibitor around the time of surgery is standard for that cancer now, because for lung cancer, melanoma, bladder cancer and triple-negative breast cancer it has become so.
Cure or control
Can immunotherapy cure cancer that has spread?
In a minority of patients, yes, in the sense that matters. Some patients treated for advanced melanoma or lung cancer stopped treatment years ago and remain free of disease, and their survival curves have flattened rather than continuing to fall. Nobody can promise this to an individual patient, because the proportion is between one in ten and one in two depending on the cancer and the markers, and because the word cure is used cautiously for a disease that can return after a decade. What can be said is that immunotherapy is the first drug treatment for spread cancer that produces long survivors in meaningful numbers, and that a patient who responds well in the first months has a real chance of being one of them.
Response is judged at three months on a scan, then every three months. A tumor that shrinks or stays still is a response. A tumor that grows on the first scan and then shrinks, which happens in a few patients as immune cells flood into it, is called pseudoprogression, and an oncologist who knows the pattern does not stop the drug on the first scan alone if the patient feels well.
What the trials show
The two trials quoted on this page were chosen because they have five years of follow-up, which is what a patient deciding today needs to know.
Median overall survival was 26.3 months for pembrolizumab and 13.4 months for chemotherapy. Five-year survival was 31.9 percent for pembrolizumab and 16.3 percent for chemotherapy. Thirty-nine patients received the full two years of pembrolizumab, and 82.1 percent of them were alive at five years (Reck and colleagues, 2021).
1
What KEYNOTE-024 tested
305 patients with lung cancer that had spread, with PD-L1 on at least half the tumor cells and no targetable mutation, randomized to pembrolizumab alone or chemotherapy. Two thirds of the chemotherapy group later received a checkpoint inhibitor when their cancer grew, and pembrolizumab still doubled survival at five years.
2
What CheckMate 067 tested
945 patients with advanced melanoma randomized to nivolumab plus ipilimumab, nivolumab alone or ipilimumab alone. The combination gave the highest survival at the cost of the most side effects, with 59 percent of patients having a serious one, and the choice between the combination and nivolumab alone is made patient by patient.
Lung cancer, high PD-L1. Single checkpoint inhibitor without chemotherapy. Five-year survival roughly doubled.
Lung cancer, lower PD-L1. Checkpoint inhibitor with chemotherapy. Survival improved in every PD-L1 group in later trials, less so as the marker falls.
Melanoma. Combination for patients fit enough to take it and for disease in the brain. Single drug for older or frailer patients. Both give a five-year survival above 40 percent.
Reading the numbers
Two things sit inside the figure of 31.9 percent. One, that most patients with this cancer still die of it within five years, and that a treatment which doubles survival is not a treatment which cures most people. Two, that the patients who reached the two-year mark on treatment mostly stayed well, so that the first year decides a great deal and the response at three and six months is the thing to watch. Early progression means a change of treatment. Response means treatment for up to two years, then watching.
The course
An infusion of thirty to sixty minutes, every two, three, four or six weeks, for up to two years. Between infusions, ordinary life.
Before the first dose. Pathology reviewed and the markers tested, PD-L1 at least, microsatellite status and a broader panel where the cancer calls for it. Blood tests for thyroid, liver, kidney, sugar and cortisol, which serve as the baseline for the side effects below. A review of autoimmune disease, organ transplants and steroid use, which change the decision.
Each cycle. A blood test the day before. A short consultation. The infusion in the day unit, with observation for an hour after the first dose and less thereafter. No hair loss, no nausea as a rule, no drop in blood counts.
Every three months. A CT or PET, compared with the last, and the decision to continue.
At two years, or earlier. Treatment stops in patients whose cancer has responded and stayed controlled, and the follow-up scans continue. Stopping early after a complete response is under study and is offered in some cancers.
How long does it take to work?
Weeks to months. Some patients feel better within the first cycle as tumor-related symptoms ease, and some scans at three months already show shrinkage. Others take six months to respond, and a few show growth first. This is slower than chemotherapy at its best and faster than it looks, because a response, once it comes, tends to last.
Bring every medicine, including steroids, and say whether you have thyroid disease, colitis, rheumatoid arthritis, lupus, psoriasis or any transplant, because these change the risk and sometimes the choice.
Keep a note of new symptoms between cycles, especially diarrhea, cough, rash, headache, fatigue beyond the usual, and thirst, and report them at the next visit or sooner.
Carry the card the unit gives you, which tells any doctor anywhere that you are on a checkpoint inhibitor and what to do about it.
1
Traveling on treatment
A patient from abroad receives the first two or three cycles here, with the baseline tests and the first scan, and then continues at home with the same drug on the same schedule if an oncologist there can give it, or returns every three to six weeks. Both patterns are common, and the oncologist here reviews every scan either way.
Side effects
Different from chemotherapy in kind and in timing. Fewer patients have them. The ones who do can have them anywhere in the body.
What are immune-related side effects?
Inflammation of a healthy organ by the immune system that the drug has unleashed. The skin, with a rash or itch, in a third of patients. The bowel, with diarrhea that can become colitis, in a fifth on combination therapy and fewer on a single drug. The thyroid, which becomes underactive in about one in ten and needs a daily tablet, usually for life. The liver, with a rise in enzymes seen on the blood test before symptoms. The lungs, with a cough and breathlessness called pneumonitis, in a few percent, more in lung cancer patients. The pituitary and adrenal glands, joints, kidneys, eyes, nerves and, rarely, the heart. Most are mild and managed by pausing the drug and giving steroids. A minority are severe and need hospital admission, and the drug is then stopped for good.
The meta-analysis of 112 trials and 19,217 patients put the rate of death from a side effect at 0.36 percent for anti-PD-1 drugs, 0.38 percent for anti-PD-L1, 1.08 percent for anti-CTLA-4 and 1.23 percent for the combination. Deaths from myocarditis, inflammation of the heart muscle, were the most lethal, with 40 percent of reported cases fatal, and deaths occurred early, at a median of 14.5 days for the combination and 40 days for single drugs (Wang and colleagues, 2018). The first eight weeks, in other words, are the weeks to be near a hospital that knows these drugs.
The rule the unit teaches every patient is short. New symptom, same day call. Diarrhea more than four times a day, breathlessness, chest pain, severe headache, confusion or a rash that blisters means the emergency department, with the card in hand.
Do side effects mean the drug is working?
Partly. Patients who develop immune-related side effects, especially of the skin and the thyroid, respond to treatment somewhat more often than those who do not, in several analyses. The relationship is loose, and a patient with no side effects can respond fully. Nobody should hope for a side effect, and nobody should read the absence of one as failure.
Combinations and sequence
Rarely alone, now.
How immunotherapy is combined
Combined with
Why
Where it is standard
Chemotherapy
Chemotherapy kills tumor cells and exposes their contents to the immune system, and it treats the patients whose markers are low
Lung, stomach, esophageal, triple-negative breast, head and neck, small cell lung cancer
A second checkpoint inhibitor
Two brakes released together, at the cost of more side effects
Melanoma, kidney, liver, mesothelioma, some lung cancers
Targeted drugs
Drugs that block tumor blood supply or growth signals
Kidney and liver cancer, endometrial cancer
Radiation
Radiation to one site can release tumor fragments that prime the immune system, and stereotactic doses are being tested for this
Lung cancer after chemoradiation, and under study elsewhere
Surgery
Given before surgery to shrink the tumor and after to clear what remains
Lung, melanoma, bladder, triple-negative breast, some bowel cancers with microsatellite instability
The combination that helps most and the one that harms least are not the same, and a fit patient in their fifties with melanoma in the brain and a frail patient in their eighties with the same disease are offered different regimens for good reasons.
Coming from abroad
Patients travel for immunotherapy when the drug is unlicensed, unfunded or unavailable at home, when the markers have never been tested, or when the first regimen has failed and a second opinion on the sequence is needed. Months of treatment suit travel less well than a procedure does. The arrangement that works is a start here and a continuation at home, or a series of visits. Which of the two suits a given patient depends on whether an oncologist at home can obtain the same drug and is willing to give it on a protocol written elsewhere, whether the patient can bear the cost of the drug at home or only here, and whether the side effects of the first cycles have been mild enough that a patient three thousand kilometers away can be trusted to recognize the ones that matter, which is a judgment the oncologist here makes at the second cycle and revisits at the first scan. Neither arrangement is second best. The drug is the same wherever it is given, and what changes is who watches.
One coordinator from the international patients office handles arrangements from the first message through discharge and answers the same WhatsApp number once you are back home. The office works in English, Arabic, French, Russian, Serbian, Romanian and Spanish and books interpreters for other languages, and it arranges the visa invitation letter, airport transfers, daily transport and accommodation near the hospital for the patient and a companion. Halal, vegetarian and diabetic meals, a prayer room on site, and a female physician on request wherever the rota allows.
What the team needs before travel is the pathology report with the marker results, or the blocks and slides if the markers have not been tested, the staging scans as image files, the record of every treatment so far with dates and response, the list of medicines, and a note of any autoimmune disease. The medical oncology team reviews the file without charge and replies with whether immunotherapy is indicated, which drug and which combination, whether the markers need testing here, the schedule and a cost estimate.
First infusion, one hour of observation, teaching on side effects, the patient card
Same-day discharge
Week 3 or 4
Second infusion, blood tests
Patients who are continuing at home leave after this cycle with the protocol
Week 12
First scan and review, here or at home with the images sent
The decision to continue
After
Cycles every two to six weeks for up to two years, scans every three months
The oncologist here reviews every scan and answers the oncologist at home
Fitness to fly
The day after an infusion, patients can fly, unless a side effect is under investigation. The length of stay for a first visit runs from one to four weeks depending on whether the markers must be tested here and whether the second cycle is given before departure. After a serious immune-related side effect, nobody flies until it has settled and the steroid dose is stable.
Questions to settle first
Five questions. Answered in advance, they save weeks.
Has the tumor been tested for PD-L1, microsatellite instability and, where relevant, tumor mutational burden. Does the cancer have a targetable mutation, which in lung cancer means a targeted tablet comes first and immunotherapy later. What stage is it, since the evidence for early disease is newer and narrower. Is there an autoimmune disease, a transplant or a need for steroids that changes the risk. And has a checkpoint inhibitor already been given, since a second course after progression on the first rarely helps and a second opinion should say so.
Not knowing the answers puts nobody at fault. The tests happen unevenly across the world, and the file review exists to find out what is missing and to order it. Immunotherapy treats the right patient. Finding out whether you are one comes before any drug is ordered, and it counts as the first treatment decision rather than a formality.
Cost
One figure, per cycle.
It depends on the drug, the dose, the schedule, the number of cycles given here, the marker tests and the scans. Hospitals in this market quote immunotherapy per cycle, with the drug as the main component and the infusion, consultation and blood tests bundled with it, and marker testing, scans and the treatment of side effects priced separately.
Confirm what a cycle includes. Confirm how the price changes if a combination or a chemotherapy partner is added. And confirm what the management of a side effect requiring admission would cost, since that is the expense patients least expect.
References
Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Five-year survival with combined nivolumab and ipilimumab in advanced melanoma. N Engl J Med. 2019;381(16):1535-1546.
Reck M, Rodríguez-Abreu D, Robinson AG, et al. Five-year outcomes with pembrolizumab versus chemotherapy for metastatic non-small-cell lung cancer with PD-L1 tumor proportion score of 50 percent or greater. J Clin Oncol. 2021;39(21):2339-2349.
Wang DY, Salem JE, Cohen JV, et al. Fatal toxic effects associated with immune checkpoint inhibitors. A systematic review and meta-analysis. JAMA Oncol. 2018;4(12):1721-1728.
Editor's note
Written by the Biruni Hospital medical editorial team. Reviewed by Prof. Dr. Özcan YILDIZ, Medical Oncology.
