摘要
The dose at which a drug enters clinical practice is largely determined by how it was first tested in humans. In oncology, Phase I trials are designed to establish the recommended dose for efficacy testing in Phase II. For decades, the dominant method has been the “3 + 3” design: small sequential cohorts of three patients receive escalating doses, and dose escalation proceeds until predefined dose-limiting toxicity thresholds are exceeded.[1] The highest dose at which fewer than one-third of patients experience dose-limiting toxicity is defined as the maximum tolerated dose (MTD), and this dose (or one level below it) becomes the recommended Phase II dose (RP2D). The underlying logic is borrowed from cytotoxic chemotherapy: more drug kills more cancer cells, and the limiting factor is what the patient can tolerate. This logic, however, does not apply to immune checkpoint inhibitors. Anti-PD-1 and anti-PD-L1 antibodies work by releasing a brake on the immune system, not by directly killing tumour cells, and their pharmacodynamic effect saturates at doses well below what patients can tolerate. In the pivotal phase I study of pembrolizumab (KEYNOTE-001), doses of 1, 3, and 10 mg/kg were tested; no dose-limiting toxicities were observed at any level, and no MTD was identified.[2] Efficacy and safety data were indistinguishable across the tested range, with a flat exposure-response relationship spanning a fivefold dose difference. Despite this, the approved dose was set at 200 mg Q3W, chosen despite the absence of evidence demonstrating superior efficacy compared with lower doses. Nivolumab, atezolizumab, durvalumab, and ipilimumab followed a similar trajectory: no MTD was reached, yet a high dose was carried forward.[2] The result is a class of drugs approved at doses that may exceed the minimum dose required for maximal biological activity, and priced accordingly. Receptor occupancy data in melanoma show that the therapeutic threshold of 70–75% PD-1 saturation is reached at nivolumab doses one-tenth of the approved amount.[3] A 2024 pharmacokinetic analysis confirmed that Cmax, AUC, response rates, and toxicity are statistically indistinguishable between a 40 mg flat dose of nivolumab and the 240 mg standard.[4] The MTD framework, conceived for a different class of drug, has produced approved doses that are not wrong in a safety sense, but are suboptimal from an access perspective. The consequences are not abstract. Of 15,674 patients who needed immune checkpoint therapy for thoracic and head and neck cancers at a single Indian tertiary centre, only 444 (fewer than 3%) received it.[5] The barrier was not regulatory delay or physician unfamiliarity. It was price. Pembrolizumab, the world’s highest-revenue drug in 2023 and approved in over 40 indications, is also among the least accessible to populations that carry the greatest cancer burden.[5] The concept of optimal biological dose (OBD), defined as the lowest dose that achieves full pharmacodynamic effect while maintaining clinical efficacy and minimising toxicity and cost, has existed in the literature for years. It has not been systematically pursued. A growing body of Indian investigator-initiated trials is now making that pursuit consequential. Evidence Across Disease Settings Classical Hodgkin lymphoma may be the most compelling proof of concept. Near-universal 9p24.1 amplification drives constitutive PD-L1 overexpression on Reed-Sternberg cells, making this disease unusually responsive to PD-1 blockade and, in theory, receptive to lower doses than solid tumours. Available data support this hypothesis across multiple settings: a matched cohort study showed identical five-year outcomes with nivolumab 40 mg versus the standard dose at a median follow-up exceeding five years;[6] salvage data combining the same low dose with conventional chemotherapy yielded complete response rates above 90%;[7] and a phase 2 chemo-free trial in which 40 mg achieved complete PD-1 receptor occupancy sustained well beyond the standard dosing interval [Table 1].[8]Table 1: Summary of key evidence supporting low-dose immunotherapy across disease settingsThe evidence for solid tumours spans metastatic, locally advanced, and neoadjuvant settings. The DELII trial, a randomised phase II study of 500 patients with relapsed or refractory solid tumours, reported a survival benefit with nivolumab 20 mg Q2W over chemotherapy, at roughly one-twelfth the standard cost.[9] In recurrent or metastatic head and neck cancer, the same low dose, when added to oral metronomic chemotherapy, numerically outperformed the 240 mg standard from CheckMate 141.[10] In the neoadjuvant setting, low-dose nivolumab combined with standard chemotherapy is yielding pathological complete response rates and event-free survival figures comparable with international standard-dose benchmarks.[10-13] The results of most of these studies have been presented as conference abstracts, and the full peer-reviewed manuscripts remain eagerly awaited. The principle extends beyond PD-1 blockade. For teclistamab, a bispecific T-cell engager targeting BCMA in multiple myeloma, regulators have formally approved dose de-escalation from weekly to fortnightly dosing for patients achieving deep responses, halving infection rates without compromising response durability.[14,15] Real-world data from over 600 patients confirm this shift in routine practice.[16,17] Even horse anti-thymocyte globulin, a pharmacologically blunt polyclonal preparation, shows comparable response rates and five-year survival at 25 mg/kg/day compared with the standard 40 mg/kg/day in severe aplastic anaemia.[18] What Should Follow This is not a call to revise global guidelines on the strength of phase II trials and retrospective cohorts. It is a call to change the question being asked. Drug development has centred on maximum tolerated dose for decades; it has not, as a discipline, seriously pursued optimal biological dose. Although emerging evidence supports lower-dose strategies, definitive confirmation through adequately powered phase III non-inferiority studies remains necessary before widespread adoption across all indications. The recent FDA Project Optimus initiative reflects growing regulatory recognition that dose selection based solely on maximum tolerated dose may be inadequate for targeted and immune therapies. Regulatory agencies need clearer pathways to approve dose de-escalation of licensed drugs, a pathway radiotherapy has pursued systematically for twenty years and that the teclistamab experience shows is achievable for immunotherapy. India has assembled, across multiple institutions and disease types, a rigorous and internally consistent body of evidence showing that effective immunotherapy need not be expensive. Whether the global oncology community chooses to act on it is no longer a scientific question. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.