Chemotherapy With Immunotherapy in India: Which Combinations Work

This page is information, not medical advice. It does not diagnose, predict an outcome, or recommend a treatment for you. Whether a chemotherapy and immunotherapy combination suits a person depends on their tumour type, stage, biomarker results, organ function and prior treatment, none of which an article can know. Use it to ask sharper questions of an oncologist who has read your reports.

Yes, oncologists across India give immunotherapy alongside chemotherapy, and for several cancers the combination has outperformed chemotherapy alone in phase 3 trials: median overall survival of 22.0 months versus 10.6 in metastatic non-squamous lung cancer (KEYNOTE-189), 28.6 versus 16.5 months in advanced cervical cancer with PD-L1 CPS ≥ 1 (KEYNOTE-826), and a halving of recurrence risk in resected stage III dMMR colon cancer (ATOMIC, hazard ratio 0.50).

Almost none of that benefit is general, which is the part most summaries drop. Each result above belongs to a specific cancer and stage, usually with a specific biomarker result attached. So the useful question for your oncologist is a narrow one: does my tumour carry the marker that made this combination work in the trial you are quoting?

There is also a distinction the phrase "experimental immunotherapy" tends to blur. Most of the combinations below are approved standard of care, not experiments. A smaller set is genuinely investigational and reachable only inside a registered trial or a compassionate-access application. Those routes differ in cost and in paperwork, and it is worth establishing which one is actually being proposed.

Key points

• Chemotherapy plus a checkpoint inhibitor (pembrolizumab, nivolumab, atezolizumab or durvalumab) is now first-line standard care in several cancers, not an experiment.
• Biomarker testing comes before the treatment decision, not after. PD-L1 CPS, dMMR/MSI-H, EGFR and ALK status each inform whether a combination is likely to help, and for some indications they set formal eligibility. For others, such as early triple-negative breast cancer, benefit held across PD-L1 levels.
• A 2026 Tata Memorial trial showed three oral drugs plus ultra-low-dose intravenous nivolumab roughly doubling one-year survival in recurrent or metastatic head and neck cancer, at a reported drug cost of about $230 a month.
• Toxicity varies by regimen rather than rising uniformly. Some combinations raise severe adverse events sharply; TMC-I below lowered them against its comparator. Immune-related effects can begin months after the last dose, so who monitors you matters as much as which drug you get.
• Genuinely experimental combinations are accessed through the Clinical Trials Registry - India (CTRI) or a compassionate-use application, not by asking a private hospital to prescribe them.

Why the two are given together

Chemotherapy and immunotherapy are often described as a simple stacking of two attacks. The reasoning is more specific than that, and understanding it explains why the combination helps in some cancers and does nothing in others.

Checkpoint inhibitors do not attack tumours. They remove a brake. Many tumours display PD-L1 on their surface, which binds the PD-1 receptor on T cells and switches them off, so the immune system recognises the cancer and then declines to act. Pembrolizumab and nivolumab block PD-1; atezolizumab and durvalumab block PD-L1. Either way the brake comes off and existing T cells resume work.

That only helps if there are T cells that recognise the tumour in the first place, which is where chemotherapy contributes. Cytotoxic drugs kill cancer cells in ways that spill tumour proteins into the surrounding tissue, giving the immune system material to recognise. Some chemotherapy agents also deplete regulatory T cells that suppress immune responses locally. So the chemotherapy is not only shrinking the tumour; it is making the tumour visible.

This explains the biomarker pattern. PD-L1 expression predicts benefit in some indications and not others: in several lung cancer regimens the combination improved survival across PD-L1 levels, which is part of why the chemotherapy is given alongside rather than the checkpoint inhibitor alone. A tumour driven by a single mutation such as EGFR tends to carry few mutations overall, presents less for the immune system to recognise, and responds better to a drug aimed at the mutation itself. A dMMR tumour, which cannot repair DNA mismatches, accumulates a very large number of mutations and so displays a great deal of abnormal protein, which is why dMMR status predicts unusually strong immunotherapy benefit and why the ATOMIC result was so large.

It also explains the toxicity. A brake released across the whole body is not released selectively at the tumour, so T cells become more active against healthy tissue too. Thyroid, bowel, skin, liver, lung and pituitary are the common sites. The toxicity is not a side effect in the incidental sense; it is the same mechanism working in the wrong place.

Regimens with phase 3 evidence, by cancer type

Regimen	Cancer	Headline result	Timing
Pembrolizumab + pemetrexed/platinum	Metastatic non-squamous NSCLC	Median OS 22.0 vs 10.6 mo; 5-yr OS 19.4% vs 11.3%	First-line
Pembrolizumab + carboplatin/paclitaxel	Metastatic squamous NSCLC	Median OS 17.2 vs 11.6 mo; 5-yr OS 18.4% vs 9.7%	First-line
Pembrolizumab + chemo before surgery, pembrolizumab alone after	Resectable stage II to IIIB NSCLC	5-yr OS 64.6% vs 53.6% (HR 0.74)	Perioperative
Pembrolizumab + chemotherapy	Early triple-negative breast	pCR 64.8% vs 51.2%; 5-yr OS 86.6% vs 81.7%	Before and after surgery
Nivolumab + XELOX/FOLFOX	Gastric, GEJ, oesophageal adenocarcinoma	Median OS 14.4 vs 11.1 mo at PD-L1 CPS ≥ 5 (HR 0.70)	First-line
Pembrolizumab + paclitaxel/platinum ± bevacizumab	Advanced cervical, PD-L1 CPS ≥ 1	Median OS 28.6 vs 16.5 mo (HR 0.60)	First-line
Durvalumab + gemcitabine/cisplatin	Advanced biliary tract	Median OS 13.0 vs 11.4 mo; 3-yr OS 15.0% vs 7.6%	First-line
Atezolizumab + carboplatin/etoposide	Extensive-stage SCLC	Median OS 12.3 vs 10.3 mo (HR 0.70)	First-line, then maintenance
Atezolizumab + mFOLFOX6	Resected stage III dMMR colon	3-yr DFS 86.3% vs 76.2% (HR 0.50)	Adjuvant
TMC-I (methotrexate, celecoxib, erlotinib, ultra-low-dose nivolumab)	Recurrent/metastatic head and neck	Median OS 10.3 vs 6.2 mo; 1-yr OS 46% vs 23%	First-line

OS = overall survival. DFS = disease-free survival. pCR = pathological complete response. CPS = combined positive score.

Lung cancer, non-small cell

For metastatic disease without an EGFR or ALK alteration, pembrolizumab added to platinum chemotherapy is standard. The regimen splits by histology: pemetrexed and platinum for non-squamous disease, carboplatin and paclitaxel for squamous. Five-year follow-up of KEYNOTE-189 and KEYNOTE-407 showed the survival gain holding rather than fading.

For resectable stage II to IIIB disease, KEYNOTE-671 gives four cycles of pembrolizumab with chemotherapy before surgery and up to 13 cycles of pembrolizumab alone afterwards, with five-year event-free survival of 49.9% against 26.5%. Nivolumab with chemotherapy before surgery is an alternative, with five-year survival of 65% versus 55% in CheckMate-816.

The EGFR and ALK caveat is not a technicality. A 2024 meta-analysis in Frontiers in Immunology of perioperative chemo-immunotherapy in resectable NSCLC found no statistically significant event-free-survival improvement in the never-smoker and EGFR-positive subgroups. That is one setting and one endpoint, and subgroup analyses are not definitive, so read it as a reason to get the molecular result before a regimen is chosen rather than as a general verdict. For tumours carrying an EGFR or ALK alteration, treatment guidelines direct the first drug at that alteration, and the evidence for adding a checkpoint inhibitor in that group is weak. Whether that applies to a given patient is for the treating oncologist to judge on the full pathology.

Triple-negative breast cancer

KEYNOTE-522 established pembrolizumab with chemotherapy before surgery, continuing afterwards for up to nine cycles, as standard for stage II to III triple-negative disease. Pathological complete response reached 64.8% against 51.2%, and overall survival improved with a hazard ratio of 0.66. Benefit held across PD-L1 subgroups, which is unusual in this field and means PD-L1 status does not gate eligibility here the way it does in gastric or cervical cancer.

Gastric, gastro-oesophageal junction and oesophageal adenocarcinoma

CheckMate 649 paired nivolumab with XELOX or FOLFOX. The gain concentrates at higher PD-L1 expression: median overall survival 14.4 versus 11.1 months at CPS ≥ 5, narrowing to 13.8 versus 11.6 months across everyone randomised. Ask for the CPS number itself rather than a yes/no, because the figure decides how much benefit is plausible. Low neutrophil counts (15%) and anaemia (6%) were the common serious toxicities, so expect frequent blood counts.

Cervical cancer

KEYNOTE-826 produced the largest absolute survival gain on this page: median overall survival 28.6 versus 16.5 months at PD-L1 CPS ≥ 1, and 29.6 versus 17.4 months at CPS ≥ 10. Grade 3 or higher events occurred in 82.4% of the pembrolizumab arm against 75.4% on chemotherapy alone. Adverse event rates are not comparable between different trials and populations, so read that against this trial's own control arm rather than against other rows of the table. Both halves of the trade-off belong in the conversation.

Biliary tract cancer

TOPAZ-1 added durvalumab to gemcitabine and cisplatin for unresectable or metastatic bile duct and gallbladder cancer. Median survival moved from 11.4 to 13.0 months, a modest shift, but three-year survival roughly doubled from 7.6% to 15.0%. Biliary cancers also carry actionable targets such as FGFR2 and IDH1, so ask about genomic sequencing alongside this regimen rather than instead of it.

Small cell lung cancer

IMpower133 combined atezolizumab with carboplatin and etoposide for extensive-stage disease, lifting median survival from 10.3 to 12.3 months, and the combination has been approved since March 2019. Four cycles of the combination are followed by atezolizumab alone until progression.

Colon cancer, dMMR

The ATOMIC trial randomised 712 patients with completely resected stage III dMMR colon cancer, 355 to FOLFOX with atezolizumab and 357 to FOLFOX alone. The combination arm had six months of chemotherapy with concurrent atezolizumab, then a further six months of atezolizumab by itself. Three-year disease-free survival was 86.3% against 76.2%, a hazard ratio of 0.50 for recurrence or death. The FDA approved atezolizumab for this indication in 2026 on the strength of the trial.

Three qualifiers are usually dropped when this result is quoted. The endpoint is disease-free survival, not overall survival, and overall survival data remain immature. It applies to resected stage III disease with confirmed dMMR, a minority of colon cancers. And grade 3 to 4 adverse events rose to 84.1% from 71.9%.

Head and neck cancer, and why the Indian trial matters

Most of the evidence above comes from trials run largely outside India, priced for health systems that are not India's. The TMC-I trial from Tata Memorial Centre is the exception. It enrolled patients with platinum-sensitive, recurrent or metastatic head and neck squamous cell carcinoma treated with palliative intent.

Among 422 patients with recurrent or metastatic head and neck cancer, 211 per arm, oral methotrexate, celecoxib and erlotinib plus 20 mg of intravenous nivolumab every three weeks was compared against paclitaxel and carboplatin. Median overall survival was 10.3 months against 6.2, one-year survival 46% against 23%, and objective response 53.4% against 24.1%. Grade 3 or higher events were lower on the experimental arm, 37.1% against 47.5%. Reported drug cost was about $230 a month, against more than $1,000 for standard checkpoint inhibitor dosing.

Note the regimen is four drugs, only three of them oral, and the nivolumab dose is a fraction of standard.

The comparator matters when reading this. Paclitaxel and carboplatin is not the first-line standard in the United States, where first-line pembrolizumab is, and survival in both TMC-I arms sits below what that regimen achieves there. That does not undo the result. It makes it a cost-constrained option rather than a new global standard.

Where these treatments are delivered in India

Capability varies more than marketing suggests. The useful question is which of four things you need next: a biomarker result, a registered trial slot, toxicity infrastructure, or a workable price.

Centre	Type	What it is built for	City
Tata Memorial Centre	Government, trial site	India's largest registered trial portfolio; ran TMC-I	Mumbai
HCG Cancer Centre	Private network	In-house tumour genomics and molecular profiling	Bengaluru
Apollo Proton Cancer Centre	Private tertiary	JCI-accredited toxicity and critical-care infrastructure	Chennai
Rajiv Gandhi Cancer Institute	Non-profit tertiary	Documented multidisciplinary tumour board and written second opinions	Delhi
Cancer Institute (WIA), Adyar	Non-profit charitable	Subsidised care and home monitoring between cycles	Chennai
Sahyadri Hospitals	Private network	Itemised cost estimates and financial counselling	Pune
Pi Cancer Care	Private single-site	Diagnostics, chemotherapy and biologics under one roof	Hyderabad

Tata Memorial Centre runs a large registered trial portfolio spanning phase I to phase III, with round-the-clock laboratory and critical-care cover, and it ran the TMC-I trial above. Some protocols are open only at registered trial sites. The cost is process: formal referral, waiting, and eligibility screening you may not pass.

HCG Cancer Centre treats molecular profiling as the first step rather than an add-on, flagging dMMR and MSI-H status before a plan is fixed. Testing before a regimen is chosen is what the trial eligibility criteria above assume. Where that testing happens is a separate question, and many centres send samples out to accredited laboratories.

Apollo Proton Cancer Centre is India's first JCI-accredited cancer hospital, with organ-based management teams and on-site endocrinology and critical care. Its relevance to this page is not the proton equipment but the speed with which severe immune toxicity can be caught and treated.

Rajiv Gandhi Cancer Institute runs complex cases through a standing tumour board across medical, surgical and radiation oncology, pathology and radiology, and has a documented written second-opinion process. That matters if you want a formal consensus on file before starting an unusual regimen.

Cancer Institute (WIA) in Adyar has treated patients regardless of ability to pay for over seven decades, and operates one of the few formal home-monitoring programmes in the country, checking patients between infusion cycles for early signs of immune toxicity.

Sahyadri Hospitals provides itemised, upfront cost estimates through a dedicated financial counselling desk rather than bundled billing that hides the immunotherapy line. Note that the TMC-I result above belongs to Tata Memorial, not to Sahyadri; low-dose approaches are a general strategy, and whether any given hospital can reproduce a trial regimen is a question to ask directly.

Pi Cancer Care in Gachibowli, Hyderabad is a private hospital founded by medical oncologist Dr Bharat Patodiya, who holds MBBS, MD and DNB Medical Oncology qualifications and, according to his own practice site, subspecialty certificates in lung and gastrointestinal cancer from the University of Zurich and in breast cancer from Ulm, Germany. Its own materials describe a daycare facility and diagnostic services. We have not been able to confirm on-site surgical or radiation oncology there, so ask which services are delivered on site and which are referred out. We have also not found registered trials listed at the practice, so if a trial place is what you need, check CTRI for any site you are considering, including this one. Directory listings disagree about his years of practice and current affiliations, so verify registration with the relevant State Medical Council or the National Medical Commission and confirm the rest with the hospital.

Biomarker testing comes first

Choosing a combination before testing is guesswork dressed as treatment. Ask for the tests that match your cancer:

• PD-L1 expression, reported as CPS or TPS, for lung, gastric and cervical cancer. Ask for the number, not a positive/negative.
• dMMR or MSI-H status for colorectal and gynaecological tumours.
• EGFR and ALK status for non-small cell lung cancer, which can rule the combination out in favour of targeted therapy.
• Hormone receptor and HER2 status for breast cancer.

Turnaround depends on the test and the laboratory. A dMMR or MSI-H panel is generally quicker than comprehensive next-generation sequencing, but ask your laboratory for its own current figure rather than relying on a general one. In fast-moving disease that wait has to be planned deliberately with your oncologist, because starting the wrong regimen before results arrive can cost more than the delay.

Reaching a genuinely experimental combination

If what you are looking for is investigational rather than approved, there are three routes and they are not interchangeable.

A registered clinical trial. Search the Clinical Trials Registry - India (CTRI), which is the statutory registry for trials conducted in India and the one most likely to list Indian sites that international databases miss. ClinicalTrials.gov is worth searching in parallel with India set as the location filter, but CTRI is the primary record here. Then approach the institutional site directly; trial coordinators, not outpatient desks, handle eligibility.

Compassionate or expanded access. Where no options remain and you do not meet trial criteria, the treating hospital can apply through the manufacturer and, in most cases, the Central Drugs Standard Control Organisation. It is slow, decided case by case, and not guaranteed.

Off-label private prescription. Legal but usually uninsured, and it places the whole cost and most of the risk on you without the monitoring structure a trial provides.

Switching between trials mid-treatment is possible but awkward, since protocols are tied to specific consent documents and site agreements, generally requiring formal exit and fresh screening. Raise it with your trial coordinator before you need it.

Cost, insurance and schemes

Get a written estimate for the full course rather than the first cycle. Then work through the levers:

• Manufacturer and hospital patient-assistance programmes.
• Insurance confirmation in writing. Most Indian policies exclude treatment classified as experimental or investigational, even at a reputable hospital, while an approved indication such as atezolizumab for dMMR stage III colon cancer is more likely covered. A verbal assurance is not coverage.
• Ayushman Bharat and state scheme eligibility, confirmed with the hospital's counselling desk, including how it interacts with other funding.
• Biosimilars, where available, and lower-dose protocols such as TMC-I where clinically appropriate.

No figure on this page is a quote. Costs move with regimen, cycle count, complications and city.

Toxicity, and who is watching

The mechanism that makes these combinations work also makes them dangerous. Reactivating the immune system produces organ-specific inflammation that can begin during treatment or months after the final dose. Across the trials above, the combinations that extended survival also increased immune-related adverse events against their own control arms.

Agree before the first infusion on what gets reported immediately: rash or itching, diarrhoea or abdominal pain, new cough or breathlessness, fever, unusual fatigue.

Some of these need emergency care rather than a message to the clinic. Breathlessness at rest, chest pain, fever during a period of low blood counts, persistent vomiting, confusion or reduced urine output all warrant going to an emergency department the same day and telling the staff there that you are on a checkpoint inhibitor. Do not wait for the next scheduled appointment, and do not treat a suspected immune reaction with over-the-counter steroids on your own. Thyroid, adrenal and pituitary dysfunction are among the more common endocrine complications and can be permanent even when caught. People with autoimmune disease, a solid organ or allogeneic transplant, or pregnancy need careful review before starting at all.

Get the monitoring plan in writing, including who to call at 2am and which hospital will admit you.

If the combination stops working

In advanced disease treated with palliative intent, progression at some point is common, and it helps to know the shape of what follows before it arrives. This section does not describe the perioperative and adjuvant regimens above, which are given with curative intent.

Progression on chemo-immunotherapy is not interpreted the same way as progression on chemotherapy alone. Checkpoint inhibitors occasionally produce pseudoprogression, where a tumour appears larger on imaging because immune cells have flooded into it rather than because the cancer has grown. It is uncommon, but it is the reason an oncologist may repeat imaging after a short interval instead of switching treatment immediately. Ask whether that is what is being considered, rather than assuming hesitation.

Where progression is real, the next step depends on what has already been used. Re-biopsy is often worth discussing, because tumours change under treatment and a marker absent at diagnosis can appear later, or a targetable mutation can emerge. A second line may be a different chemotherapy backbone, a targeted drug if sequencing finds a target, a trial, or a shift in goals toward symptom control.

Palliative and supportive care is active treatment aimed at symptoms and function, and it can run alongside anticancer treatment rather than replacing it. Specialist guidelines recommend introducing it early rather than only once treatment stops. Asking about it is not giving up, and a team that treats the question as defeat is telling you something about the team.

Limitations of the evidence on this page

Most of these trials enrolled few Indian patients, and conclusions about efficacy and toxicity are being transferred across populations with different genetics, comorbidity patterns and baseline nutrition. TMC-I is the clearest exception.

The TMC-I figures come from an ASCO 2026 presentation and secondary reporting of it rather than a full peer-reviewed paper, and conference data sometimes shifts on publication. Secondary sources also disagree on its grade 3 adverse event rates: The ASCO Post and Targeted Oncology both report 37.1% against 47.5% with P = .036, while at least one other outlet reports slightly lower figures. This page uses the better-corroborated pair and flags the disagreement rather than hiding it.

Median survival is a midpoint, not a prediction. Half the patients in these trials did worse than the number quoted. And none of these results tell you what will happen to one person, which is the only question that actually matters to you.

Questions worth asking

1. Which biomarker result qualifies me for this regimen, and what is the actual number?
2. Which trial supports it, and did that trial enrol patients at my stage and with my marker?
3. Is this approved for my indication in India, or investigational? If investigational, which route are we using?
4. What is the expected response rate for my subtype, not the overall trial average?
5. Which toxicities will be monitored, how often, and who do I call outside clinic hours?
6. What is the plan if the immunotherapy has to stop early?
7. What does the full course cost, in writing, and what has my insurer confirmed in writing?

A centre that answers all seven clearly has thought about your case.

Second opinions and tumour boards

A combination regimen is a reasonable trigger for a second opinion, and in India the practical obstacle is usually logistics rather than cost.

There are two different things people mean by a second opinion. One is another oncologist reviewing your reports and giving a view, which can often be done remotely if you can send pathology, biomarker results and imaging. The other is a tumour board, a standing panel across medical, surgical and radiation oncology with pathology and radiology, reviewing the case together. For an unusual regimen or a borderline eligibility question, the board carries more weight than any single additional opinion, because disagreement surfaces in the room rather than between two reports you then have to reconcile yourself.

Ask for the outcome in writing. A documented, dated consensus is useful if an insurer later queries the regimen, and it gives the treating team something concrete to work against.

What to bring, in every case: the histopathology report, the full biomarker panel including the actual PD-L1 score rather than an interpretation, recent imaging with the images and not only the report, a list of every treatment already given with dates and cycle counts, and current blood counts and organ function tests. A second opinion given without the biomarker numbers is an opinion about a different patient.

One caution. Seeking opinions indefinitely is itself a decision, and in aggressive disease delay carries its own cost. Two well-prepared opinions are usually more useful than five rushed ones.

Verifying any centre, including the ones listed here

Marketing language in Indian private oncology runs well ahead of capability, and the terms that matter are not protected. Questions that separate the two:

• Is the tumour board a standing scheduled meeting with minutes, or an informal corridor conversation described as one?
• Is biomarker testing done in-house or sent out, and to which accredited laboratory? Ask for NABL accreditation.
• For a trial: what is the CTRI registration number? A centre that cannot produce one is not running a registered trial.
• Who manages immune toxicity out of hours, and is on-site endocrinology and critical care available every day or only on weekdays?
• Will the full-course cost estimate be itemised, with the immunotherapy line separated from bundled charges?

Frequently asked questions

Does immunotherapy replace chemotherapy? Rarely, in these regimens. Both are given together, and some patients continue on immunotherapy alone as maintenance after the chemotherapy course finishes.

Is "chemo plus immunotherapy" experimental? Usually not. Most combinations on this page are approved standard care for their specific indication. A smaller number are genuinely investigational, and those are reached through a registered trial or compassionate access rather than a private prescription.

Can side effects appear after treatment ends? Yes. Immune-related effects can start months after the last dose. Keep reporting new symptoms to your oncology team after treatment finishes, and tell any other doctor you see that you received a checkpoint inhibitor.

Does insurance in India cover this? Approved indications often yes, investigational use usually not. Get written confirmation from your insurer naming the drug and indication before starting.

How long does treatment last? It depends on the regimen. Courses before surgery run three or four cycles. Some regimens cap maintenance at a protocol-defined maximum, for example 35 cycles of pembrolizumab in KEYNOTE-189, while others continue until progression or unacceptable toxicity. Ask which applies to yours.

Should I get a second opinion? For any combination regimen, yes. Take your pathology report, biomarker results and scans so the second oncologist can check the marker and the regimen rather than re-run the diagnosis.

Are subcutaneous injection forms available? Subcutaneous pembrolizumab and nivolumab formulations have been approved in the United States. Availability in India changes, so ask your oncologist or hospital pharmacy rather than assuming.

Once more, plainly: nothing here is advice about your treatment. Every figure on this page is a trial result in a defined population, not a forecast for you. Take your reports to a qualified oncologist and ask the seven questions above.

---

Reviewed 6 November 2026 by the India Cancer Guide editorial team. This article has not been reviewed by an independent clinician. Every clinical claim is cited to a primary or reference source, and our sourcing policy sets out how those are chosen and what we will not publish. Cost figures, scheme eligibility and trial status all change; if you find something stale or wrong, tell us and we will correct it and date the correction.