Complete surgical removal of localized cancer cures more cancer patients than any other form of treatment, in part because this modality is generally applied as sole treatment for local, early stage disease, or tumors with limited potential to metastasize. In humans, 60% of patients who are cured of cancer are cured by surgery alone (Poston). Before this hope for cure can be realized in veterinary medicine, surgeons must have a thorough understanding of anatomy, physiology, resection, and reconstruction options for all organs, expected tumor behavior, and the various alternatives or adjuvants to surgery. Surgical oncologists should not only be good technicians (cancer carpenters), but dedicated tumor biologists. Surgery will play a role at one point or another in the management of most cancer patients. This utility of surgery may include any of the following: diagnosis (biopsy), resection for cure, palliation of symptoms, debulking (tumor cell cytoreduction), and a wide variety of ancillary procedures to enhance and complement other forms of treatment.
Surgical resection of cancer was introduced in the 16th century B.C. and remained relatively underutilized until general anesthesia (1840s), antisepsis (1860s), and pain management made aggressive resection safe and tolerable for the patient. Dr. William Halstead developed the basic principles of surgical oncology in the 1890s. The radical resection of the 20th century has increasingly been customized to meet the needs of the patient and has frequently been reduced in magnitude. Further refinements in surgery have been made possible with newer equipment (staples, endoscopy, etc.), advanced imaging, pain management, use of blood products and critical care services.
Most patients with cancer are "old." However, "old" is a relative term, and a geriatric dog or cat with normal organ function should not be denied treatment simply on the basis of age. Age has not been shown to impact tumor-related prognosis. In fact, dogs with osteosarcoma that are less than 2 years of age do worse than dogs that are more than 2 years of age after amputation alone. "Old" animals, in most instances, will tolerate aggressive surgical intervention as well or as poorly as "young" patients.
Surgery for Diagnosis
Although biopsy principles are covered in Chapter 9, it bears emphasizing that properly timed, performed, and interpreted biopsies are one of the most crucial steps in the management of the cancer patient. Not only does the surgeon need to procure adequate and representative tissue to establish a diagnosis, additionally, the biopsy must not compromise subsequent curative surgical resection or radiation field planning.
Surgery for Cure
Before a surgeon can be in a position to provide the optimal operation for the patient with cancer, the following questions need to be considered:
1. What is the histologic type, stage, and grade of cancer to be treated?
2. What are the expected local and systemic effects of this tumor type, grade, and stage?
3. Is a cure possible and at what price in terms of cosmetics and function?
4. Is an operation indicated at all?
5. What are the options for alternative or planned combination treatment?
A recurring theme in surgical management of cancer is that the first surgery has the best chance of cure. Several mechanisms for this improvement in survival have been advanced. Untreated tumors have had less chronologic time to metastasize than recurrent primary cancer. Untreated tumors have near-normal anatomy, which will facilitate operative orientation and maneuvers. Recurrent tumors may have had seeding of previously noninvolved tissue planes, requiring wider resection than would have been required on the initial tumor. If one thinks about a given cancer as resembling a crab, incomplete surgery removes the body of the crab and leaves the legs behind. The "body" of most tumors is often quiescent and hypoxic, whereas the leading edge of the tumor (legs) is the most invasive and well vascularized. Subtotal removal may actually selectively leave behind the most aggressive components of the tumor. Patients with recurrent cancer will often have less normal tissue for closure. An ill-defined negative aspect of recurrent cancer is reported to be related to changes in vascularity and local immune responses. Regardless of the mechanism, curative-intent surgery is best performed at the first operation and the surgeon should have all the necessary diagnostic information in hand when constructing a treatment plan. The increased availability of diagnostic imaging such as computed tomography (CT) has added greatly to our ability to determine the extent of a solid tumor and optimize the surgical approach. CT allows good visualization of muscle bellies/fascial planes, intraabdominal/intrathoracic organs, lymph nodes and bony detail. Three-dimensional CT reconstructions are particularly useful for planning procedures for skull-based tumors.
Advanced imaging has greatly enhanced the surgeon's ability to assess the anatomic location and extent of various cancers; however, we must remember that imaging of any kind needs to be paired with clinical palpation and assessment of mobility. Some cancers deemed inoperable by imaging are in fact mobile and operable. Leading edges of some cancer are compressed against adjacent tissue and appear more invasive. Surgeons should always take the opportunity to palpate the local tumor before or after imaging, explore the history of the tumor's growth pattern, and in many cases obtain a tissue sample (i.e., histopathology or cytology when appropriate) before declaring a mass inoperable. Positive prognostic factors typically include slow growth rate, mobile within the tissues, first attempt at surgery, discrete tumor borders, small tumor size and low-grade nature. Conversely, surgery may be less effective for the same tumor type and grade if the mass is ill-defined, recurrent, or has a recent history of rapid growth. Palpation under heavy sedation or anesthesia may suggest that resection is possible in spite of imaging findings (e.g., unilateral thyroid masses, some soft tissue sarcomas). Deeper masses such as rib and liver may appear inoperable due to inflammation or compression but not invasion of adjacent organs and structures. Lymph nodes that appear enlarged on imaging studies but cannot be easily sampled prior to surgery due to location (e.g., tracheobronchial lymph node for a lung tumor) pose a real decision-making dilemma.
The surgical oncologist must be able to assimilate all of the information and make an informed decision; however, we must also remind ourselves and our clients that there is much we do not know - i.e., incomplete margins do not necessarily mean later local recurrence (Bacon et al.) and that surgical judgment on expected local behavior and likely resection is often qualitative and is an imperfect "science."
The actual surgical technique will vary with the site, size, and stage of the tumor. The same tumor type in dogs and cats may vary in surgical approach, technique, and prognosis. Some general statements that need to be emphasized with surgical oncology are the following:
1. All incisional biopsy tracts should be excised in continuity with the primary tumor, since tumor cells are capable of growth in these wounds. Fine-needle aspiration cytology tracts are of minor, but not zero, concern, while punch biopsy tracts are of intermediate concern. With this in mind, all biopsies should be positioned in such a manner that they can be removed at surgery.
2. Early vascular ligation (especially venous) should be attempted to diminish release of large tumor emboli into the systemic circulation. This is probably only clinically meaningful for those tumors with a well-defined venous supply, such as splenic tumors, retained testicles, and lung tumors. Small numbers of cancer cells are constantly being released into the venous (and lymphatic) circulation by most tumors. Larger, macroscopic cell aggregates may be more dangerous, however, and these may be prevented from vascular escape with early venous ligation.
3. Local control of malignant cancer requires that variable margins of normal tissue be removed around the tumor. Resection of the "bad from the good" can and should be classified in more detail than radical versus conservative. Tumors with high probability of local recurrence (high-grade soft-tissue sarcoma, high-grade mast cell tumors, feline mammary adenocarcinoma, etc.) should have 2- to 3-cm margins removed in three dimensions. Tumors are not flat, and wide removal in one plane does not ensure complete excision. Fixation of cancer to adjacent structures mandates removal of the adherent area in continuity with the tumor. This is commonly seen with oral cancer that is firmly adherent to the underlying mandible or maxilla. Invasive cancer should not be peeled out, shelled out, enucleated, or curetted if a cure is expected. Many cancers are surrounded by a pseudocapsule. This capsule is almost invariably composed of compressed and viable tumor cells, not healthy reactive host cells. If a malignant tumor is entered at the time of resection, or if the margins of removal come back as incomplete from the pathologist, that procedure is often no better, therapeutically, than a large incisional biopsy. When possible, resection of the previous scar and the entire wound bed with "new" margins (never entering the old wound cavity) is indicated. One should strive for a level of dissection that is one tissue plane away from the mass. For example, invasion of cancer into the medullary cavity of a bone requires subtotal or total bone resection and not curettage.
The width of surgical margins necessary for a complete excision for a given tumor type is an ongoing debate, and our current practices are based on little objective data. As a community, we have answered most of the questions about how much tissue we can safely remove, but it will serve our patients well to determine how little extra tissue is necessary to excise and consistently achieve the same success. We must challenge recommendations that are reported in the literature if based solely on a surgeon's personal experience or opinion in the absence of objective data.
4. Tumors should be handled gently to avoid risk of breaking off tumor cells into the operative wound, where they may thrive. Copious lavage of all cancer wound beds will help mechanically remove small numbers of exfoliated tumor cells but should not replace gentle tissue handling and avoidance of entering the tumor bed.
5. If more than one malignant mass is being removed, separate surgical packs should be used for each site to avoid iatrogenic tumor cell implantation from site to site.
The aggressiveness of resection should only rarely be tempered by fears of wound closure. It is better to leave a wound partially or even in some cases completely open with no cancer than closed with residual cancer. Numerous innovative reconstructive techniques are available for closure of cancer wounds, and the surgeon is only limited by his or her ingenuity. Reliable microvascular free composite transfers of muscle and skin are somewhat hampered due to unique canine skin/muscle anatomy but are being developed.
Lymph Node Removal
Controversy surrounds the surgical management of regional lymph nodes draining the primary tumor site. As a general rule, epithelial cancers are more likely to metastasize to lymph nodes than are mesenchymal cancers. However, any enlarged regional lymph node requires staging investigation. Lymphadenopathy may be from metastasis of cancer (firm, irregular, and sometimes fixed to surrounding tissue) or from hyperplasia and reactivity to various tumor factors, infection, or inflammation. The former cause is a poor prognostic sign, and the latter may be a beneficial host response. Enlarged lymph nodes as a result of cancer metastasis and invasion are generally uniformly effaced by tumor cells and can often be diagnosed by fine-needle aspiration. Histologically, positive lymph nodes at diagnosis usually are a sign of impending emergence of systemic metastasis.
Lymph nodes should be removed under two general circumstances:
1. If the lymph node is positive for cancer and not fixed to surrounding normal tissues, it may be possible to remove the node with some therapeutic intent. Frequently, however, many lymph nodes drain a primary tumor site (e.g., oral cavity) and lymphadenectomy is incomplete. Lymph node metastasis at the time of initial diagnosis is a poor prognostic sign. However, some patients develop metastasis in a delayed fashion (1 to 2 years) after local tumor control and may benefit from lymphadenectomy. Although it is usually not practical, removal of the primary tumor, intervening lymphatic ducts, and draining lymph node has been recommended (en bloc resection). En bloc resection may be possible for a malignant toe tumor with metastasis to the popliteal lymph node, but is usually only accomplished with amputation. A mastectomy that includes the regional lymph node (e.g., glands four and five with the inguinal lymph node) is another example of en bloc resection. Few other anatomic sites are routinely amenable to this therapy. A specific instance where local lymphadenectomy may be beneficial is in removal of iliac/sublumbar lymph nodes in patients with metastatic apocrine or sebaceous gland adenocarcinomas of the perineum. Although removal of these lymph nodes is rarely curative, it may help increase the benefit of radiation therapy and should help alleviate, at least in the short term, the paraneoplastic syndrome of hypercalcemia by reducing levels of parathormone-like substances. It may also prevent or improve obstruction of the large bowel and urinary tract and regional lymphadenectomy can be repeated as necessary in the absence of overt systemic metastasis (i.e., metastasis to lungs, liver, etc.).
2. It is well established that normally sized lymph nodes may contain micrometastasis, and this point should always be explained clearly to a client. Normal-appearing lymph nodes that are known to drain a primary tumor site may be randomly sampled (biopsy or cytology) to gain further staging information. This is particularly important if adjuvant therapy decisions (irradiation or chemotherapy) would be predicated on confirmation of residual or metastatic cancer. Intrathoracic or intra-abdominal lymph nodes are perhaps most crucial, since they are not readily accessible to histologic or cytologic follow-up examination. However, in some instances lymph nodes that are not enlarged cannot be sampled safely due to proximity to vital structures (e.g., sublumbar lymph nodes at the aortic bifurcation), even under ultrasound guidance. In such cases the surgeon must educate the client about the situation and either remove the primary tumor without further knowledge of regional lymph node involvement or recommend removal of the normal-sized nodes concurrently for staging (and possibly therapeutic) purposes. For dogs with malignant anal sac tumors, this approach involves an exploratory surgery of the sublumbar nodal bed and dissection/removal of any and all nodal tissue encountered. In human medicine, the concept of the sentinel lymph node (first node to receive lymphatic flow from a primary tumor) has become important in the management of select malignancies, most notably breast cancer. Basically, the area of the primary tumor is injected with blue dye or a low dose of a radionuclide or both. The first draining node is detected visually (dye) or with a handheld gamma camera probe and removed for a frozen section. If the first node is negative for metastasis, subsequent nodal dissection often is avoided. If the sentinel node is positive, further nodal dissection is performed. Targeting the sentinel node is most valuable in an anatomic location where there is an extensive nodal bed (e.g., the human axilla) and hence numerous potential paths for regional metastasis. The benefits of such an approach in humans are readily apparent, as many patients have been spared extensive nodal resections if the sentinel lymph node has been determined to be free of cancer cells. The topic of the sentinel lymph node has only recently emerged in veterinary medicine but is gathering some momentum. Several techniques have been described, including microbubble detection via ultrasound, fluorescein and blue dye, and though it may not yet have the relevance and importance that have been demonstrated in humans (i.e., nodal networks of dogs are less complex, complications not typically encountered when all the nodes in a region are removed), there are several potential indications for its use (e.g., tumors of the head and neck) that should be investigated in clinical studies. Histologically positive lymph nodes will alter prognosis and stage and will also be informative for decisions related to postoperative chemotherapy and radiation.
Lymph Nodes Are Not Removed Under Two General Circumstances
1. Lymph nodes in critical areas (retropharyngeal, hilar, mesenteric) that have eroded through the capsule and become adherent (fixed) to surrounding tissues cannot be curatively removed without serious harm to the patient. They are best biopsied and left in situ or treated with other modalities. The occasional exception is metastasis of limb and paw tumors to prescapular and popliteal lymph nodes that can be removed with amputation (radical en bloc resection).
2. Prophylactic removal of "normal" draining lymph nodes or chains of lymph nodes (as opposed to sampling for stage) is of no benefit and may be harmful. Regional lymph nodes may in fact be the initiator of favorable local and systemic immune responses, and elective removal has been associated with poor survival in certain human cancers.
Surgery for Distant Disease
Metastasectomy for pulmonary metastasis of sarcomas has been accepted therapy in humans and dogs. Resection of liver metastasis for carcinomas (especially gastrointestinal cancers) is increasing in human oncology. As more effective adjuvant therapies evolve and minimally invasive techniques are further developed, the need for cytoreductive metastasectomy will increase.
Palliative Surgery
Palliative surgery is an attempt to improve the quality of the patient's life (pain relief or improved function) but not necessarily the length of the patient's life. This type of surgery requires careful consideration of the expected morbidity of the procedure versus the expected gain to the patient and the client. In essence, it comes down to a decision of when to discontinue therapy. One of the most difficult decisions in surgical oncology is the decision not to operate. Treatment of any kind should never be worse than no treatment.
Certain situations do exist, however, where palliative surgery may be beneficial. If an infected and draining mammary tumor in a patient with asymptomatic lung metastasis is the limiting factor in the patient's life, mastectomy may still be a logical procedure. Splenectomy for ruptured hemangiosarcoma is commonly performed but probably has little impact on long-term survival and can be considered palliative, since it will stop the immediate threat of hemorrhage.
Debulking Surgery
Incomplete removal of a tumor (planned or unplanned) is referred to as debulking or cytoreductive surgery. It is commonly performed but rarely indicated. Its theoretical indication is to enhance the efficacy of other treatment modalities. Debulking is a practical consideration prior to cryosurgery to decrease the amount of tissue to freeze and the time it will take. It may also help the treatment planning and dosimetry with certain forms of irradiation, but the improved cancer control achieved is more a result of geometric and dosimetry considerations than removal of a few logs of tumor cells. Removing 99.9% of a 1-cm tumor (1x109 cells, one billion) still leaves a million cancer cells behind. Immunotherapy and chemotherapy could theoretically be helped by tumor volume reduction (such as lymph node removal for oral melanoma with the use of a melanoma vaccine), but few well-controlled clinical trials have shown a benefit to date in veterinary medicine. A variety of soft tissue sarcomas in dogs and cats will have better local control when radiation is given adjuvantly rather than for bulky measurable disease. Amputation or limb sparing of dogs with osteosarcoma is essentially a debulking procedure and clearly requires postoperative chemotherapy for prolongation of life. If tumors are debulked with the anticipation of postoperative radiation therapy, the margins of known tumor or the operative field should be marked with radiopaque metal clips to allow proper treatment planning from radiographs. The orientation of the incision should be considered carefully if radiation therapy is possible postoperatively.
Nonsurgical, Locally Ablative Procedures
Ablative techniques to eradicate local (or metastatic) disease have a place in oncology but often suffer from sales before science and are only rarely based on evidence of outcomes. Several of these techniques will be mentioned below, and the reader is referred to accompanying references for details. Indications vary but are generally limited to "small" (<2-cm diameter) discrete lesions with some palpable or image evidence of tumor margins to try and eradicate. All of the techniques suffer from inability to assess completeness of cell kill and margin analysis after treatment. If the operator picked the correct tumor, site, and "dose" delivery, then local control may be achieved; however, monitoring for regrowth is the only way to find out. As a general rule, recurrent disease tends to be more invasive and difficult to control than the primary intervention, so the first ablative maneuver is hopefully well planned and executed. Any of the local ablative treatments require special equipment and training to properly perform. Selective tumor cell kill while sparing all normal tissue is often claimed for these techniques but is unlikely to be true.
Surgery and Chemotherapy
The combined use of chemotherapy and surgery is becoming more commonplace in veterinary oncology, and the knowledge an oncologic surgeon must possess to master the use of combination therapy is ever-expanding (e.g., tyrosine kinase inhibitors for mast cell tumors). Many chemotherapy agents will impede wound healing to some extent. In spite of this risk, few clinically relevant problems occur when surgery is performed on a patient receiving chemotherapy. General recommendations are to wait 7 to 10 days after surgery to begin chemotherapy, especially for high-risk procedures such as intestinal anastomosis. The use of intraoperative or perioperative chemotherapy is receiving increased attention and could have greater implications for wound healing. Neoadjuvant chemotherapy is also becoming more popular and in some instances may greatly facilitate excision of a solid tumor. Such an approach is common with canine mast cell tumors, but there are still unanswered questions. Which tumors would benefit? If a tumor reduces in size, should the location of the original tumor border be used to make measurements for margins or the new outer edge?
Surgery and Radiation
Theoretical advantages can be advanced for both pre- and postoperative radiation. Either way, some impairment of wound-healing potential will exist. Radiation damage to normal tissues (stem cells, blood vessels, and lymphatics) is more permanent than chemotherapy damage. As total radiation dose, dose per fraction, and field size increase, the potential complications (with or without surgery) increase. If radiation therapy is given preoperatively, surgery can be performed after acute radiation reactions have resolved (generally 3 to 4 weeks). Postoperative radiation is recommended to start immediately or after a 10- to 14-day delay. In spite of the theoretical problems, surgery can often be safely performed on irradiated tissues. Complications may occur but are not prohibitive.
The benefit of surgery and radiation is clear for some tumor types; however, in some instances the improvement in outcome over single modality therapy is controversial. One such debate is that of canine nasal tumors. Early reports did not show benefit of postoperative radiation over radiation alone. The pendulum has been swinging back and forth, however, with a recent report showing preoperative radiation to have benefit when followed by exenteration of the nasal cavity 6–10 weeks later (Adams). With the advent of conformal radiation therapy (stereotactic radiosurgery [SRS] and intensity-modulated radiation therapy [IMRT]), such approaches are even more attractive since the overlying skin or underlying mucosa in the case of nasal tumors can be spared from the full effects of the radiation, and concerns about incisional healing thereby diminished. Where in the past surgeons would be reluctant to operate in a radiation field, these more focused forms of radiation therapy will hopefully allow the surgeon to operate with fewer wound-healing complications and create novel treatment plans that combine radiation and surgery for select tumors. Radiation side effects are greatly diminished with a more conformal approach, and this in turn makes clients more willing to have their pets undergo radiotherapy.
Access to more sophisticated radiation techniques (i.e., SRS, IMRT) is rapidly increasing in veterinary radiation facilities and with this development a new paradigm is emerging in veterinary oncology. For example, bone sarcomas in locations not amenable to limb-sparing surgery can now be treated with curative intent therapy (Farese), and treatment protocols for large solid tumors previously deemed non-resectable are currently being investigated. Such treatments may be done in one very high dose with SRS or fractionated into several moderately high doses over several days with stereotactic radiotherapy (SRT). Combinations of SRS/SRT and surgery are also being explored. While these new treatment options represent great advances, familiar challenges remain. In the case of dogs with appendicular osteosarcoma, fracture may still occur after SRS or SRT and while some of these cases are amenable to surgical stabilization, healing of the fracture does not usually occur due to the effects of radiation on bone healing. Thus, the role of the surgeon continues to evolve in the management of cases treated with radiation.
Miscellaneous Oncologic Surgery
Veterinary surgeons are being called upon increasingly to facilitate the management of cancer patients. The placement of long-term vascular access catheters for delivery of fluids, chemotherapy, or anesthesia and pain relief agents has become commonplace and ports are routinely placed to aid in the evacuation of malignant thoracic effusions (e.g., mesothelioma). Operative placement of various enteral and parenteral feeding tubes is also commonly performed. The comprehensive veterinary oncology team now calls on the discipline of interventional radiology to manage/palliate certain malignancies (e.g., urethral and ureteral stenting for malignant urinary obstructions).
Surgeons and radiotherapists may work together for the operative exposure of nonresectable cancer so that large doses of irradiation may be delivered to the tumor or tumor bed after exclusion of radiosensitive tissues. Surgical intervention for oncologic emergencies such as intractable pain, bleeding, pathologic fracture, infection, and bowel perforation or obstruction may also arise.
Equipment advances are facilitating tumor excisions (e.g., LigaSure for splenectomy and liver masses) and laparoscopic and thoracoscopic evaluation of body cavities for staging is increasingly being performed on animals. Cancer resections with this technique are also on the rise and, given the decreased morbidity associated with these techniques, veterinary surgeons may feel more comfortable performing surgery in the face of advanced disease for certain solid cancers (e.g., thoracoscopic removal of a solitary metastatic lung nodule at the time of amputation for an appendicular osteosarcoma). A few other examples of minimally invasive surgery performed in companion animals for cancerous diseases include prostatic biopsy, thoracoscopic pericardectomy for heart-based tumors, laparoscopy-assisted intestinal biopsy, liver biopsy, pancreatic biopsy, splenectomy (Holak), and adrenalectomy (Mayhew). Further definition of appropriate case selection and increased access to equipment and training will ultimately expand these techniques as the rising popularity of minimally invasive surgery to the pet-owning public is driving this technology.
It is clear that surgery will be the mainstay of cancer treatment in veterinary medicine for many years to come. It is also clear that just because a surgical procedure is possible, this is not the best reason to do it. It was not long ago that surgical resection of the external genitalia was routine treatment for dogs with transmissible venereal cell tumors. It is now recognized that chemotherapy alone is curative in over 90% of dogs, and surgery is needed for biopsy only. Simple versus radical mastectomy in the dog and humans does not influence survival for most mammary gland tumors, but more aggressive surgery may indeed be beneficial in the cat. More surgery is not always better surgery. Long-term followup of well-staged and graded tumors with defined surgical technique and margins is necessary to demonstrate the true value of any operation. A great deal of progress in surgical technique and surgical thinking needs to take place before the use of surgery can be optimized.
It is hoped that a better understanding of expected tumor biology and more precise staging methods (molecular diagnostics, angiograms, ultrasound, CT scans, MRI, PET/CT, etc.) will facilitate more precise surgical operations to be performed. Surgical technique will continue to improve and undergo refinements, but until surgeons become biologists, the big breakthroughs will be slow in coming. Surgeons should be investigating the influence of anesthesia, infection, immune function, blood transfusions, growth factors, oncogenes, and cytokines, to name a few, on the outcome of our patients. In spite of these anticipated advances in technology and biology, the most difficult aspect to learn is surgical judgment. Biology is King; selection of cases is Queen, and the technical details of surgical procedures are the Princes and Princesses of the realm who frequently try to overthrow the powerful forces of the King or Queen, usually to no long-term avail, although with some temporary apparent victories.
The incorporation of the Veterinary Society of Surgical Oncology (VSSO) is further evidence of the maturity of surgical oncology in veterinary medicine and the American College of Veterinary Surgeons (ACVS) is now serving as a credentialing body for Surgical Oncology Fellowship programs. The VSSO should provide for well-defined surgical oncology trials and enhance teaching, training, research, and service in this critical area of oncology.