Of all forms of gastrointestinal malignancy, adenocarcinoma of the pancreas is associated with the worst survival. Most patients die within a year after establishment of the diagnosis. Epidemiological studies, underlined by animal experiments, reveal that cigarette smoking and alcohol consumption contribute to the increase of the incidence of pancreatic cancer in industrialized countries; hereditary cancer syndromes are implicated in less than 8% of malignant lesions of the pancreas [1].

In spite of the progress in surgical treatment, resulting in increasing resection rates and a decrease in treatment-related morbidity and mortality, the true figures of cure are even today below 3% [2]. In this article present knowledge about dissemination roots of pancreatic cancer is re-evaluated in the light of results of current treatment protocols including surgery, adjuvant chemotherapy and radiotherapy, and neoadjuvant treatment modalities.

Dissemination Pattern of Pancreatic Cancer: Limitation of Surgical Treatment

The international classification systems of pancreatic cancer [International Union Against Cancer (UICC), American Joint Cancer Committee (AJCC), Japanese Pancreatic Society (JPS)] rely on tumor size, lymph node (LN) involvement, stage of infiltration into surrounding tissues, and presence of metastasis. The cardinal rule in improving the prognosis proved to be complete tumor removal in patients undergoing oncological resection; in recent published series, the absence of residual tumor is associated with an increased chance of survival. The prognosis of patients undergoing a resection of pancreatic cancer is determined by the state of lymph node (LN) metastasis, tumor size, invasion of blood vessel walls, and number of units of blood transfused during surgery and in the early postoperative course. Furthermore, factors of tumor biology have also been found to have a prognostic effect; these include DNA ploidy status, cell differentiation, and the absence of aberrations of oncogenes and suppressor genes (Table 1). A patient without LN metastases and a tumor size < 2 cm, without vessel wall involvement or distant metastases, has a significant survival benefit after a R0 resection [9].

Table 1. Ductal cancer of the pancreas: factors of tumor biology favorably influencing survival after resection.
Full size table

Unfortunately preoperative staging is unreliable regarding the presence and extent of LN involvement. The N-factor can only be clarified after surgical resection by histological examination. The reliability of LN negativity is related to the size of the operative specimen sent for pathological examination, which is determined by the extent of the lymphadenectomy performed. Patients undergoing a standard Kausch-Whipple resection for pancreatic head cancer without clearance of the more distant LN stations are at risk of being classified as false-negative for N-status; metastasis to LN in the hilum of the liver, the inter-aortocaval spaces, and the left side of the superior mesenteric artery is frequent [10]. Among patients with pancreatic head cancer, histological examination techniques have demonstrated that about 30% of the LN in the inter-aortocaval space are cancer infiltrated [11]. The posterior hepatic LN, the LN of the hepatoduodenal ligament, and the posterior pancreatic head LN drain primarily to the LN in the inter-aortocaval space below the left renal vein. These roots are the major origin of the thoracic duct [12]. In a collective series of patients with a small tumor (< 2 cm), LN metastases were discovered in one third [13]. Careful histopathomorphological evaluations of cancer disseminations have demonstrated that in cancer stage I and II, lymph vessels surrounding the head of the pancreas are cancer cell infiltrated (Table 2).

Table 2. Dissemination pattern of pancreatic cancer.
Full size table

Using molecular biological methods like reverse transcriptase polymerase chain reaction (RT-PCR) or immunostaining, a new dimension of micrometastasis has been objectified. With the higher sensitivity of these molecular-biological methods, up to 60% of LN previously seen as microscopically free of cancer cells showed micrometastasis by RT-PCR (Table 3). In bone marrow specimens micrometastasis was found in 36% to 65% of the patients investigated (Table 4). Among patients in cancer stage I and II (UICC) undergoing surgery, 46% had positive immunostaining for cancer cells in the bone marrow. Using RT-PCR techniques 13 of 17 investigated patients showed micrometastases in the liver; some but not all of these patients later developed metastatic liver disease macroscopically [24].

Table 3. Micrometastasis in HE-negative lymph nodes.
Full size table
Table 4. Pancreatic cancer micrometastasis.
Full size table

Present knowledge about cancer cell dissemination early in the course of pancreatic disease, including UICC stages I and II, explains the observed frequency of recurrence in more than 95% of the patients undergoing surgical resection.

Nerve plexus invasion outside of the pancreas has been observed in 43%–72% of patients, occurring most often in the right plexus mesentericus II and the nerve plexus around the superior mesenteric artery [29] (Table 5).

Table 5. Nerve plexus invasion outside of the pancreas.
Full size table

Cancer of the Uncinate Process Bears the Worst Prognosis

The uncinate process arises from the embryological ventral bud of the pancreas. The development of a malignancy within this area of the pancreas is frequent. Because of the lack of early symptoms like jaundice, patients with cancer of the uncinate process have a poor prognosis. Vessel wall involvement occurs early in the course of the disease, and is present in almost all patients with advanced uncinate cancer due to the proximity of the malignant lesion to the mesenteric vessels. At the time of diagnosis, liver metastases are present macroscopically in more than one third of these patients [30]. Lymph node involvement has been observed in 20% in the inter-aortocaval spaces and in 65% on both sides of the superior mesenteric artery and the mesentery of the small bowel. Because of the dissemination pattern of cancer of the uncinate process, after oncologic resection patients have shorter survival chances than patients with pancreatic head cancer arising from the dorsal bud of the pancreas [9].

Pancreatic Cancer Treatment: Achievements of Surgery

Although surgery offers a low cure rate, it is also the only chance for cure. Ductal pancreatic cancer in an advanced stage is diagnosed in ≥ 95% of patients; potentially curative resection can be performed only in about 10%–15% of them. Contraindications for resective surgery are the presence of liver metastases or distant metastases, peritoneal seedings, circular tumor infiltration into mesenteric vessel walls, and extension of the tumor into the mesentery of the jejunum or the mesocolon transversum. Lymph node enlargement near and remote from the primary tumor is not considered a criterion for non-resective management.

Major contributions to improved surgical treatment results are the reduction of hospital mortality and treatment in high-volume centers (Table 6). After the first article reporting a large series of Whipple resections without any deaths [39], the hospital death rate in experienced centers is now < 3%–5%. Postoperative morbidity has also decreased dramatically as a result of standardization of surgical techniques with well-defined steps of tissue clearance. With both standard and extended oncological resections, the early postoperative complication rate in experienced institutions is < 30%–40%. Standardized surgical techniques for suturing anastomoses have led to a decrease in severe local complications such as pancreatic fistula, intraabdominal bleeding, and leakage at the site of intestinal anastomoses, minimizing local septic complications. After a Kausch-Whipple resection, 60%–75% of the patients are discharged from the hospital between the 8th and 15th postoperative day [40]. Present knowledge about the survival benefits yielded by a more extended tissue clearance does not support an oncological extended Kausch-Whipple resection in pancreatic head cancer [41] [42]. However, the extent of tissue resection remains to be determined on the basis of knowledge about the degree of dissemination, even in cancer at stages (UICC) I and II. Presumably patients with local lymph node involvement [node stage 13, 17 (JPS)], and with an N0-LN status of the N2 LN and no vessel and nerve-plexus involvement, may gain a long-term survival benefit from extended oncological resection [43]. More than half of the few patients observed to survive > 5 years after pancreatic cancer resection had an advanced cancer stage with positive lymph nodes, serosal involvement, and vessel wall involvement (Table 7). Extended resection including resection of vessel wall can be performed without increased hospital morbidity and mortality (Table 8). Portal vein and/or superior mesenteric vein resection in patients with limited vessel wall infiltration results in a downstaging of the cancer and therefore a survival benefit; in about 50% of the patients in which the surgeon considered the vessel wall to be infiltrated, there was actually an adherence of the tumor but microscopically no cancer infiltration into the adventitia (Table 8).

Table 6. Pancreatic carcinoma: hospital mortality after pancreaticoduodenectomy—results of low- and high-volume hospitals.
Full size table
Table 7. Long-term survival after resection: observed 5-year survivors.
Full size table
Table 8. Pancreatic head resection: portal (PV) and superior mesenteric vein (SMV) resection.
Full size table

R0 Resection Fails to Improve Long-term Survival

The stage of residual tumor R0–R2 is determined by histological examination of tissue of the resection margins of the pancreas, common bile duct, stomach, and duodenum, respectively. In most recent published prospective trials, R0 resection results in an increase of survival in comparison to patients with a residual tumor; after R1 or R2 resection, no long-term survivors are reported. The achievement of R0 resection is determined by the extent of tissue dissection. Considering present knowledge of dissemination patterns of pancreatic cancer, it is a mistake to identify R0 resection with absence of residual tumor. Many R0 series published after standard tissue resection of the pancreatic head cancer by means of a Kausch-Whipple procedure are hampered by a failure to include remote cancer cell–positive tissues in the operative specimen—e.g., N2 LN, nerve plexus, and perivascular tissues. Cancer recurrence after so-called resection with curative intent is frequently the consequence of cancer cell–positive tissues left behind (Table 9).

Table 9. Recurrence after R0 resection.
Full size table

More than 95% of the patients undergoing surgical resection are in an advanced stage of cancer. R0 resection established by histological examination of resection margins is reliable only in cases in which a full tissue specimen is histologically investigated. However, using a standard Kausch-Whipple resection in pancreatic head cancer, the N2 LN as well the nerve plexus on the right side of the aorta are not part of the operative specimen. Cancer infiltration in N2 LN is present in 30%–60% of stage II and III cancers. Recurrence of the cancer develops in 40% of the patients within 6 months and in 60%–80% within 12 months of surgical resection. The progression-free period varies between 8 and 12 months (median) [64, 65, 66, 67]. In one third of the patients undergoing R0 resection, liver metastasis is the first sign of recurrence. In these patients liver metastases have been overlooked during surgery.

Regarding long-term survival after R0 resection, only 3%–16% of the patients from selected series survived 5 years or more (Table 7). Comparing the survival times after standard and extended resection of pancreatic head cancer no significant long-term survival benefit results from extended R0 resection [41, 42, 68].

Survival Statistics and the Definition of Treatment End Points in Pancreatic Cancer Surgery

Long-term survival ≥ 5 years is observed in a very small group of pancreatic cancer patients, contradicting the published 5-year actuarial survival rates of 20%–45% among resected patients. Over the period of 65 years of resective cancer treatment Gudjonson claims that not more than 300–350 individuals with observed and well documented 5-year survival have been reported in the international literature [69]. The Kaplan-Meier calculations of survival result in misleading survival figures if, in the subgroup of patients treated by surgical resection, the hospital deaths and the patients lost to follow-up are excluded. Publication of actuarial survival figures should be considered as unacceptable without information on the total number of patients in the study group, the number of observed (actual) survivors, and definition of the subset of patients followed after resection.

A few end point evaluations have been conducted in pancreatic cancer treatment trials. Evidence from studies shows that surgical resection as well as chemotherapy can prolong survival and improve quality of life in advanced pancreatic cancer. The number of patients who benefit from treatment is, however, still limited. The assessment of clinical benefit from surgical or medical cancer treatment should be based on several end points beyond actuarial survival only. Besides the well-defined actual survival, the observed median survival has been employed as a simple, reliable, and well-evaluated criterion. In addition, progression- free survival has recently been introduced in surgical series as one of the most appropriate primary end points from which to measure the treatment benefits.

In the palliative setting, as in most cases of pancreatic cancer, treatment-related survival prolongation is of greatest importance; single or combined chemotherapy as well as multimodality regimen have shown only modest effectiveness, with objective response rates of most protocols in the range of 10%–20%. Response evaluations rely on the application of sensitive staging methods. Response rates to chemotherapy are not strongly correlated to a survival benefit. Doubts have been raised regarding whether response rate is an independent prognostic factor for survival [70]. Clinical benefit response has been introduced as an additional end point to evaluate the efficacy of chemotherapeutic agents; a combination of improvements in pain (reductions in pain intensity and/or analgesic requirements), performance status, and weight gain is used to objectify clinical benefit [71]. Clinical benefit response, however, can underestimate the effects of chemotherapy because it does not include the assessment of other symptoms. Further, it can overestimate the results of chemotherapy, as it does not properly assess the side effects. The assessment of quality of life (QOL) aspects using the European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Index C30 (QLQ-C30) [72] or a standard functional assessment of hepatobiliary cancer treatment (FACT-Hep) [73] has now been incorporated in cancer trials to accomplish the more traditional assessment. In advanced pancreatic cancer QOL was significantly better in patients receiving chemotherapy than in those receiving best supportive care. The level of QOL after pancreatic cancer resection is strongly determined by the presence (low QOL) or absence (high QOL) of cancer recurrence.

In pancreatic cancer treatment trials reporting results after oncologic resections, the primary end points to evaluate treatment efficacy are median survival (months), actual survival at 1 to 5 years, and progression-free survival (months). Reporting only actuarial or cumulative survival figures is inadequate. In series with multimodality treatment, clinical benefit response as well as QOL measurements using EORTC, QLQ-C30 are of greatest importance in addition to observed survival data.

Adjuvant Treatment Improves Survival

The limits of surgery are defined by low resectability rates and the biology of the disease, which determines the patients’ poor prognosis even after R0 resection at an average median survival of 11 to 24 months (Table 10). Multimodality treatment concepts have applied radiotherapy and chemotherapy either alone or in combination before, during, or after surgical resection. Two carefully controlled prospective series conducted by the Gastrointestinal Tumor Study Group (GITSG) (1985, 1987) [74, 75] were able to extend survival in patients with ductal pancreatic cancer with postoperative 5-fluorouracil, doxorubicin, and mitomycin C (FAM)-chemotherapy versus surgery alone. Patients receiving adjuvant chemotherapy had a median survival of 20 months versus 11 months among patients who had surgical treatment alone. In the GITSG studies the survival benefit persisted; after 10 years 19% of the patients treated with radiochemotherapy were still living, whereas no patients from the surgery-alone group survived. In a European adjuvant treatment trial (EORTC-GITCCG) patients with curatively resected pancreatic head cancer were randomized into two groups, one receiving 40 Gy + 5-FU for radiosensitization and the other receiving surgery alone. The median survival in the two groups was 23.5 months for the radiosentized patients versus 19.1 months for those who received surgery alone. Although the difference was not significant, the adjuvant treatment seemed to be effective in pancreatic head cancer patients (Table 10).

Table 10. Pancreatic cancer: adjuvant treatment: results of controlled clinical trials.
Full size table

In a prospective case control study, Yeo et al. [76] confirmed the benefits of postoperative adjuvant radiochemotherapy; the Baltimore trial resulted in an increase in median survival to 19.5 months versus 13.5 months in the surgery-alone group. The European Study Group of Pancreatic Cancer recently finished the largest randomized controlled clinical trial evaluating the benefits of adjuvant chemotherapy using 5-FU/folinic acid. The median survival of the patients who had oncological cancer resection and postoperative 5-FU/folinic acid treatment was 19.7 months in 238 patients versus 14 months in the surgery-alone group [78]. The median survival was highly significant; a significant difference was observed for the 2-year actuarial survival. The results of the three prospective controlled studies, two of them randomized series, evaluating the benefits of adjuvant treatment with large patient allocation and sufficient observation periods demonstrated a significant benefit of 6–10 months with regard to the median survival time.

Regional adjuvant chemotherapy has been shown to improve the survival time in studies comparing intraarterial chemotherapy using celiac artery infusion versus historical controls. The median survival time improved after regional adjuvant chemotherapy to 21 months versus 19.3 months in historical controls [79]. Celiac artery infusion exposes the rest of the pancreas and the upper abdominal organs, and particularly the liver, via the hepatic artery and splenic–portal vein, to chemoactive drugs. The preliminary data deriving from prospective clinical trials, comparing intraarterial regional adjuvant chemotherapy with historical controls, reveal that disease progression was significantly reduced in the liver. In fact, tumor recurrence occurred either locally or in the peritoneum, but it occurred in the liver in < 20% of the cases (Table 11) [79, 81].

Table 11. Ductal pancreatic cancer: options to prevent/reduce occurrence of liver metastases using regional adjuvant chemotherapy.
Full size table

The effect of radiation alone is still under discussion. Local disease control and longer survival were achieved by Zerbi et al., who administered a high dose of intraoperative radiation therapy (IORT) [82]. In contrast, the local relapse rate in the GITSG patient group receiving 40 Gy radiotherapy was 33% versus 47%–55% in the control group. In the Mayo trial, patients were treated with 54 Gy; only 9% had a local relapse, but the disease progression in the liver was 52%, resulting in a limited survival benefit for the adjuvant-treated patients in spite of the improved local disease control [83].

A positive effect of intraoperative radiotherapy after resection as a single treatment modality has not been unanimously confirmed; however, in a prospective controlled clinical trial, a median survival time of 13 months was achieved in the treated patients versus 8 months in the control group [82]. The combination of extended radical resection and IORT improved the actuarial 5-year survival rate to 29% versus 0% [84]. Taking the results of the prospective trials together—although they are on a lower level controlled—IORT in combination with oncological resection leads to a significant reduction in local recurrence and a prolongation of survival to 12.8–16 months in comparison to a survival of 7–8 months in control groups [85, 86].

Impact of Neoadjuvant Treatment

A protocol for neoadjuvant, multimodal treatment of pancreatic cancer is not yet established. Results from uncontrolled, prospective mono-institutional series applying radiochemotherapy to patients with pancreatic cancer stage II and III (UICC) resulted in a frequency of downstaging of 15%–30% and a resection rate of the downstaged patients between 50% and 83% [87, 88, 89]; the median survival rates of these patients ranged 15–32 months. The use of preoperative chemoradiation is supported by the following considerations: (1) The goal of neoadjuvant treatment is downstaging of the patient and, in combination with an oncological resection, increasing the chances of survival. A certain percentage of potentially unresectable tumors are downstaged to enable surgical resection. (2) Radiation therapy is more effective on well-oxygenated cells that have not been devascularized by surgery. (3) Pretreatment before surgery may prevent implantation and dissemination of tumor cells at laparotomy. (4) Patients with evidence of disseminated disease on re-staging after chemoradiation will not be subjected to unnecessary laparotomy. (5) Delayed postoperative recovery will not affect the delivery of multimodality therapy as it does in one third of the patients receiving adjuvant chemotherapy.

In recent published controlled clinical trials comparing historical and prospective control groups, the frequency of downstaging was observed to be between 13% and 45% [88, 89]. Oncological resection after neoadjuvant radiochemotherapy resulted in a median survival between 15 and 32 months [84, 85, 86]. Patients with UICC stages II and III pancreatic cancer are candidates for neoadjuvant treatment. Neoadjuvant multimodal treatment including radiotherapy with 54 Gy and chemosensitization using 5-FU/folinic acid or gemcitabine had a survival benefit after resection in comparison to non-resected patients of the same cancer stage. Between 10% and 25% of the patients with resectable cancer are downstaged [86].

During neoadjuvant chemotherapy, disease progression occurs in 15%–25% of the patients with the appearance of liver metastases or peritoneal carcinosis. These patients are spared a laparotomy. After neoadjuvant radiochemotherapy, patients who were not considered candidates for surgical resection because of adherence of the cancer to the wall of the portal vein or the superior mesenteric vein show a separation between tumor and vessel wall [89]. Downstaging in this group of advanced pancreatic cancer patients resulted in a survival benefit after oncological resection. Neoadjuvant radiochemotherapy also resulted in a decrease in the frequency of cancer-positive resection margins. Finally, after neoadjuvant radiochemotherapy and surgical resection no increase in postoperative complications has been reported [89].

Summary

The prognosis of patients suffering pancreatic cancer who undergo surgical resection is determined by the state of lymph node metastasis, invasion of blood vessel walls, infiltration of extrapancreatic nerve plexus, and the degree of micrometastasis into the surrounding tissues and remote organs. More than 95% of such patients are in an advanced stage of cancer. Major contributions of surgery to improve treatment results are reduction of hospital morbidity and mortality. In high-volume centers the hospital mortality is considered to be < 5%. Regarding long-term survival after R0 resection, < 10% of patients in selected surgical series were observed surviving 5 years and more without cancer recurrence. Long-term survival is observed in a very small group of patients, contradicting the published 5-year actuarial survival rates of 20%–45% of resected patients. Kaplan-Meier analyses of survival results in misleading survival figures in the subgroup of patients treated by surgical resection if patients who died in the hospital and those lost to follow-up are excluded from the calculations. R0 resection alone fails to improve long-term survival. Besides the well-defined actual survival, other reliable criteria to measure treatment results including observed median survival, progression-free survival, treatment-related survival prolongation, and quality of life data are accepted and reliable criteria to assess treatment results. Adjuvant treatment improves survival after oncological resection. The survival benefit by applying chemotherapy or radiochemotherapy in an adjuvant setting has been demonstrated to be 6–10 months in terms of median survival time. After oncological resection of pancreatic cancer each patient should be offered adjuvant treatment. A neoadjuvant treatment protocol for pancreatic cancer is presently not established. However, after neoadjuvant radiochemotherapy about 15% of downstaged patients have a survival benefit in combination with an oncological resection.

Résumé.

En dépit des progrès réalisés dans le traitement chirurgical, essentiellement une augmentation du taux de la résécabilité et une diminution de la morbidité et de la mortalité, le taux de cure après résection pour cancer du pancréas reste, même aujourd’hui, en dessous de 3%. La dissémination locale du cancer du pancréas limite le taux de survie à court terme (< 3 ans) et à long terme des patients ayant eu une résection. Selon une évaluation histologique, moins de 15% des patients ayant eu une résection R0 étaient pN0, avec plus de 60% qui avaient un envahissement lymphatique; > 50% des patients ont une infiltration extrapancréatique plexique nerveuse. En fait, les ganglions HE négatifs s’étaient montrés cancéreux lorsque la coloration PCR ou l’immuno-coloration ont été utilisées au niveau de ces ganglions HE négatifs. Les cancers du petit pancréas ont un très mauvais pronostic en raison d’une absence de signes précoces; l’envahissement vasculaire, fréquent, est habituellement précoce; il y a souvent également des métastases hépatiques. La chirurgie est souvent la seule chance de cure, quoique faible. Le diagnostic de cancer canalaire est fait dans plus de 95% des cas à un stade avancé; une résection potentiellement curatrice peut être réalisée chez seulement environ 10–15% des patients. Le pas important de la chirurgie pour améliorer les résultats du traitement a été la réduction de la morbidité chirurgicale, c’est-à-dire les complications postopératoires locales et systémiques ainsi qu’une diminution de la mortalité hospitalière arrivant en-dessous de 3–5%. Selon les résultats des essais prospectifs les plus récents, la réalisation d’une résection R0, comparée à une résection laissant de la tumeur en place, augmente la survie. Cependant, ceci n’est pas toujours vrai. Dans beaucoup de séries de résection R0 publiées après résection standard des cancers de la tête du pancréas selon le procédé de Kausch-Whipple, on a noté l’absence de tissus comprenant des cellules cancéreuses dans les pièces de résection, c’est-à-dire au niveaux des ganglions lymphatiques N2, des plexus nerveux et des tissus périvasculaires extrapancréatiques et rétropancréatiques. La récidive après des résections soi-disant R0 avec intention de cure est souvent la conséquence du tissu cancéreux laissé en place. La survie à long terme (> 5 ans) peut être observée dans un groupe de patients extrêmement petit, avec des taux de survie actuarielle à 5 ans de 20–45% chez les patients réséqués. L’évaluation des bénéfices cliniques provenant d’un traitement chirurgical or médical devrait être basée sur plusieurs points, pas seulement sur la survie actuarielle. Les publications concernant la survie actuarielle sans information en ce qui concerne le nombre de patients en survie réel (actuel), la définition d’un sous-groupe de patients suivis après résection et le nombre total de patients dans le groupe d’étude dans les essais thérapeutiques du cancer du pancréas peuvent prêter à confusion: on a en effet besoin de critères de jugement plus convaincants pour évaluer l’efficacité thérapeutique après résection oncologique, tels que la médiane de survie (en mois), la survie actuelle à 1–5 ans et la survie sans progression de la maladie (en mois). Dans les séries de traitement multi modalité, l’évaluation de la réponse clinique ainsi que la qualité de vie utilisant des instruments de mesure comme les questionnaires EORTC ou QLQ-C30 sont également très importants en plus des chiffres de survie. Le traitement adjuvant améliore la survie après résection oncologique. Cependant, les bénéfices à court et à long terme après chimiothérapie adjuvante en cas de résection R0 ou R1-2 sont jusqu’à présent sans conviction à partir des essais cliniques contrôlés. Le bénéfice de survie par la chimiothérapie ou par la radio chimiothérapie est de l’ordre de 6–10 mois en ce qui concerne la médiane de survie. Après résection oncologique d’un cancer du pancréas, chaque patient a droit à un traitement adjuvant. Il n’existe pas, à l’heure actuelle, de traitement néoadjuvant bien établi pour cancer du pancréas.

Resumen.

Los progresos registrados en el tratamiento quirúrgico del adenocarcinoma de páncreas han propiciado un incremento en el número de resecciones y un descenso en las tasas de morbi-mortalidad inherentes a la intervención quirúrgica; sin embargo, el número real de pacientes curados no supera el 3%. La rápida diseminación del cáncer al espacio retropancreático es responsable de la escasa supervivencia, a corto (< 3 años) y largo plazo, de los pacientes resecados. Utilizando criterios histológicos, menos del 15% de los pacientes sometidos a una resección curativa (R0) pertenecen al estadio pN0 y más del 60% presentan una linfangitis carcinomatosa; cerca del 50% de los pacientes muestran infiltración carcinomatosa en los plexos nerviosos extrapancreáticos. Los ganglios linfáticos negativos con la tinción de hematoxilina-eosina (HE) resultan positivos, con micrometástasis, si se utilizan otras técnicas como la RT-PCR o la inmunotinción. El cáncer del proceso uncinado tiene muy mal pronóstico pues cursa inicialmente de forma asintomática; con frecuencia, la invasión de las paredes vasculares se produce muy precozmente y las metástasis hepáticas también. El tratamiento quirúrgico constituye la única terapéutica efectiva, pero sólo en pocos casos tiene carácter curativo. En más del 95%, el adenocarcinoma ductal pancreático se diagnostica en estadios avanzados y una resección, potencialmente curativa, se realiza sólo en un 10–15% de los pacientes. El progreso de la técnica quirúrgica ha reducido exclusivamente la morbilidad, p. ej. las complicaciones locales o sistémicas postoperatorias, disminuyendo la mortalidad intrahospitalaria por debajo del 3–5%. En estudios prospectivos recientes se ha demostrado un aumento de la supervivencia a corto plazo en pacientes con resecciones radicales curativas R0, con respecto a aquellos en los que persiste un resto tumoral. Sin embargo, las resecciones R0 no han mejorado la supervivencia tardía. Estudios casuísticos publicados han demostrado que tras resección R0 (efectuando la duodenopancreatectomía estándar a lo Kausch-Whipple), persisten restos tumorales tales como: micrometástasis en los ganglios linfáticos N2, nidos de células cáncerosas extrapancreáticas en tejido retroperitoneal, perivasculares o a lo largo de los plexos nerviosos. La recidiva, tras la así llamada resección R0 curativa, se debe, con frecuencia, a estos restos neoplásicos abandonados, por desapercibidos, en la operación estándar. Supervivencias > 5 años se constatan sólo en un reducido grupo de pacientes, lo que contrasta con las cifras publicadas basadas en la curva actuarial que alcanza hasta el 20–45% de los pacientes resecados. Para averiguar la variable predefinida que permite cuantificar los efectos del tratamiento quirúrgico o médico del cáncer de páncreas, han de utilizarse otros criterios de valoración distintos a la supervivencia actuarial. Las publicaciones de supervivencia actuarial sin cifras que informen sobre el número de casos actualmente vivos ni que expliciten el número de pacientes revisados tras la resección, en relación con el número total de los estudiados, conducen a conclusiones erróneas. En los ensayos sobre el tratamiento del cáncer de páncreas los resultados, por lo que a la eficacia de las resecciones oncológicas se refiere, han de basarse en criterios de valoración principales tales como: supervivencia media (en meses), supervivencia actual al 1–5 años, y el curso evolutivo de la supervivencia sin enfermedad (en meses). En series en las que se aplican tratamientos multimodales han de valorarse además la respuesta clínica al tratamiento así como la calidad de vida de los pacientes evaluados mediante tests tales como el EORTC, y el QLQ-C30. El tratamiento adyuvante aumenta la supervivencia tras resecciones oncológicas. Sin embargo, en la actualidad no existen estudios controlados que demuestren fehacientemente, que el tratamiento con quimioterapia adyuvante, tras resecciones R0 y R1–2 sea beneficioso para los pacientes ni a corto ni largo plazo. Lo único que se ha demostrado es que la administración de quimio o radio-quimioterapia adyuvante prolonga en 6–10 meses la vida de los pacientes con respecto a la supervivencia media. Tras una resección oncológica por cáncer de páncreas cada paciente a de ser sometido a un tratamiento adyuvante, pero hasta el momento no se ha podido definir el tratamiento neoadyuvante más idóneo para el cáncer de páncreas.