2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
The conduct of randomized trials in surgical oncology, although highly appealing in concept, may be problematic, especially in complex settings where the skills, experience, and clinical judgment of individual surgeons and their institutions may vary greatly.
Maurie Markman, MD
The primary role of surgery in the management of malignant disease is indisputable. This modality has been central to the fundamental success of oncology therapeutics from the earliest days of cancer treatment, when essentially the only valid option for cure or palliation was surgical removal of a mass or masses (“solid tumors”), until today, when ever-improving imaging techniques and experience have strikingly refined the extent of such surgical resections. All one has to do is consider the remarkable evolution of the standard-of-care curative approach in the management of breast cancer to appreciate the magnitude of changes in surgical oncologic strategies; the field has moved from the historical Halsted radical mastectomy to the far less morbid lumpectomy followed by external beam radiation.
In recognition of the foundational and impressively evolving role of surgery in cancer management, the American Society of Clinical Oncology (ASCO) named “refinement of surgical treatment of cancer” as its Advance of the Year for 2020.1 The ASCO report highlighted examples of highly relevant changes in the surgical approach that have fundamentally and positively affected outcomes and quality of life for patients with cancer.
The administration of neoadjuvant cytotoxic chemotherapy in several clinical settings has permitted subsequent surgical intervention to remove a primary malignancy, as in breast cancer, or safely and effectively cytoreduce residual macroscopic tumors, as in ovarian cancer. Increasingly effective targeted therapy in certain clinical settings, such as renal cell cancer and melanoma, has permitted systemic therapy to become the primary treatment modality, with surgery employed to remove partially responsive tumor masses or residual disease unresponsive to an antineoplastic regimen. Initial systemic or local radiation therapy may also permit resection of a locally advanced primary malignancy, as in pancreatic cancer. Finally, the delivery of systemic therapy or local radiation therapy may also permit less extensive but equally effective curative resections, such as in the case of limb sarcomas in children and adults.
Of course, these highly relevant clinical advances are critically informed by the conduct of prospective clinical trials, or the publication of analyses of prospective or retrospective experiences at single or multiple centers with expertise in surgical oncology.
Differences in Modalities
Although surgical, radiation, and systemic treatment strategies are all important in cancer management, there are fundamental and pragmatic differences between these concepts within the clinical trial domain and ultimately in routine clinical practice. Although the inherent biology of an individual cancer may ultimately trump the ability of any modality to favorably influence the natural history of that malignancy, the impact that the oncology specialist has on cancer-specific and overall outcomes may vary greatly depending on the modality.
Consider an example―admittedly somewhat extreme― that highlights the point of this commentary. If an internationally recognized expert in the antineoplastic drug management of “cancer A” orders standard-of-care “treatment B” for “patient C,” there is absolutely no reason to believe that the biological or clinical response to treatment, such as shrinkage of the tumor mass or masses, reduction in tumor markers, or improvement in cancerrelated symptoms, would be different if the same regimen were ordered by a first-year medical oncology trainee during the first day of her/his fellowship.
In striking contrast, no one would make such an outlandish claim that outcomes would be equivalent for a surgical oncologist with several decades of experience in managing a specific cancer type compared with a surgical resident during her/his first year or day of training. In this situation, the combined experience of a given surgical team and hospital in dealing with such complex surgeries also should be considered.
Again, although these scenarios may be a bit extreme, the point is clear: Cancer surgery, especially complex procedures undertaken in advanced or difficult circumstances as well as innovative procedures such as robotic surgery, demands knowledge, particular skills, and experience to optimize the opportunities for the most favorable possible clinical outcomes.
As a result, the conduct of randomized trials in surgical oncology, although highly appealing in concept, may be problematic, especially in complex settings where the skills, experience, and clinical judgment of individual surgeons and their institutions may vary greatly.
It is common to hear surgeons described by their surgical and nonsurgical colleagues as being “aggressive” or “conservative” in their basic approach to management within the operating room. Without passing judgment on any surgical philosophy, we can consider the question of whether such perspectives, along with the previously noted characteristics of skill and experience, might affect the results of a cancer trial in the surgical domain.
An Attempt at Objective Criteria
In an interesting and provocative report, Japanese investigators have attempted to assess objectively the relationship between certain outcomes and the surgeon’s experience and ability to undertake complex surgery for gynecologic cancers.2 They evaluated operative time and total blood loss in a consecutive population of ovarian cancer surgeries (N = 271) performed only by gynecologic oncologists and divided the procedures into 2 categories of complexity. The investigators found that approximately 50 operations were required before proficiency was achieved in high-complexity cases. Not surprisingly, a lower number of total cases was required for an individual surgeon to be classified as being proficient in less-complex surgeries.
Of course, there is nothing magical about 50 cases or any other number of procedures that would define the necessary experience for a given surgeon to be able to successfully undertake a case and achieve an optimal clinical outcome, including operative and postoperative morbidity and mortality.
However, it is important to appreciate the unique relevance of such skills and experience in the design and interpretation of surgical clinical trials in oncology and in reports of outcomes from individual institutions and groups. Further, this heterogeneity in surgical providers may help at least partially explain different conclusions that have been reached regarding the impact of innovative surgical strategies on clinical outcomes, such as minimally invasive radical surgery in the management of cervical cancer.3,4